WO2022110797A1

WO2022110797A1 - Calibration method and apparatus, electronic device, and storage medium

Info

Publication number: WO2022110797A1
Application number: PCT/CN2021/102895
Authority: WO
Inventors: 陈龙泉; 刘余钱
Original assignee: 上海商汤临港智能科技有限公司
Priority date: 2020-11-30
Filing date: 2021-06-29
Publication date: 2022-06-02
Also published as: CN112362054B; CN112362054A

Abstract

Provided are a calibration method, a calibration apparatus, an electronic device, and a computer-readable storage medium. The calibration method comprises: when a traveling device is traveling, acquiring pose data of a combined inertial navigation device on the traveling device and three-dimensional point cloud data collected by a radar sensor on the traveling device (S101); determining pose data of the radar sensor on the basis of external parameter information indicating a coordinate system conversion relationship between the radar sensor and the combined inertial navigation device and the pose data of the combined inertial navigation device (S102); determining, on the basis of pose data of the radar sensor at different time points and three-dimensional point cloud data collected at the corresponding time points, stitched point cloud data obtained from stitching three-dimensional point cloud data of a target region (S103); and adjusting the external parameter information on the basis of the stitched point cloud data, and obtaining target external parameter information (S104).

Description

一种标定方法、装置、电子设备及存储介质A calibration method, device, electronic device and storage medium

相关申请的交叉引用CROSS-REFERENCE TO RELATED APPLICATIONS

本专利申请要求于2020年11月30日提交的、申请号为202011370652.6、发明名称为“一种标定方法、装置、电子设备及存储介质”的中国专利申请的优先权，该申请以引用的方式并入本文中。This patent application claims the priority of the Chinese patent application filed on November 30, 2020 with the application number of 202011370652.6 and the invention titled "a calibration method, device, electronic device and storage medium", which application is by reference Incorporated herein.

技术领域technical field

本公开涉及数据处理技术领域，具体而言，涉及一种标定方法、装置、电子设备及存储介质。The present disclosure relates to the technical field of data processing, and in particular, to a calibration method, an apparatus, an electronic device, and a storage medium.

背景技术Background technique

无人驾驶车辆也叫自动驾驶车辆，是指在没有人工参与的情况下，能够感知周围环境并进行自主驾驶的车辆。在自动驾驶车辆中，一般通过设置于自动驾驶车辆上的定位传感器来进行定位，定位的准确性将直接影响其后期行驶的可靠性。An unmanned vehicle, also called an autonomous vehicle, refers to a vehicle that can perceive the surrounding environment and drive autonomously without human intervention. In an autonomous vehicle, positioning is generally performed by a positioning sensor set on the autonomous vehicle, and the accuracy of the positioning will directly affect the reliability of its later driving.

由于单个定位传感器或多或少存在一些自身的局限性，因此当下采用多个传感器来进行综合定位。在基于多个传感器进行综合定位时，需要将不同传感器采集的位姿数据转换到同一个坐标系，具体是通过表征不同传感器之间的相对位置关系的外部参数信息进行转换的。在无人驾驶车辆的行驶过程中，外部参数信息可能会发生变化，需要对其进行标定，来获取更准确的外部参数信息。Since a single positioning sensor has more or less its own limitations, multiple sensors are currently used for comprehensive positioning. When performing comprehensive positioning based on multiple sensors, it is necessary to convert the pose data collected by different sensors into the same coordinate system, specifically, the conversion is performed by external parameter information representing the relative positional relationship between different sensors. During the driving process of the driverless vehicle, the external parameter information may change, and it needs to be calibrated to obtain more accurate external parameter information.

发明内容SUMMARY OF THE INVENTION

本公开实施例至少提供一种标定方案。The embodiments of the present disclosure provide at least one calibration solution.

第一方面，本公开实施例提供了一种标定方法，包括：获取在行驶设备行驶过程中，所述行驶设备上的组合惯性导航设备的位姿数据、以及所述行驶设备上的雷达传感器采集的三维点云数据；基于表征雷达传感器与所述组合惯性导航设备之间坐标系转换关系的外部参数信息、以及所述组合惯性导航设备的位姿数据，确定所述雷达传感器的位姿数据；基于所述雷达传感器在不同时间点的位姿数据、以及在相应时间点采集的三维点云数据，确定对目标区域的三维点云数据进行拼接后得到的拼接点云数据；基于所述拼接点云数据，对所述外部参数信息进行调整，得到目标外部参数信息。In a first aspect, an embodiment of the present disclosure provides a calibration method, including: acquiring pose data of an integrated inertial navigation device on the traveling device during a traveling process of the traveling device, and collecting data collected by a radar sensor on the traveling device The three-dimensional point cloud data; based on the external parameter information representing the coordinate system conversion relationship between the radar sensor and the integrated inertial navigation device, and the pose data of the integrated inertial navigation device, determine the position and attitude data of the radar sensor; Based on the pose data of the radar sensor at different time points and the 3D point cloud data collected at the corresponding time points, determine the splicing point cloud data obtained by splicing the 3D point cloud data of the target area; based on the splicing point Cloud data, adjust the external parameter information to obtain target external parameter information.

本公开实施例中，在通过预先确定的外部参数信息确定出雷达传感器在不同时间点的位姿数据后，这样可以结合雷达传感器在不同时间点采集的三维点云数据，确定出不同时间点下表征目标区域的三维点云数据，在外部参数信息不再准确的情况下，不同时间点下表征目标区域同一位置点的坐标值不再相同，使得拼接点云数据中的点变多，基于此，可以基于拼接点云数据对外部参数信息进行调整，从而不断优化外部参数信息，直至得到准确度较高的目标外部参数信息。In the embodiment of the present disclosure, after the position and attitude data of the radar sensor at different time points are determined through the predetermined external parameter information, the three-dimensional point cloud data collected by the radar sensor at different time points can be combined in this way to determine the position and attitude data at different time points. For the 3D point cloud data representing the target area, when the external parameter information is no longer accurate, the coordinate values of the same point representing the target area at different time points are no longer the same, resulting in more points in the spliced point cloud data. Based on this , the external parameter information can be adjusted based on the spliced point cloud data, so as to continuously optimize the external parameter information until the target external parameter information with high accuracy is obtained.

在一种可能的实施方式中，所述基于所述雷达传感器在不同时间点的位姿数据、以及在相应时间点采集的三维点云数据，确定对目标区域的三维点云数据进行拼接后得到的拼接点云数据，包括：基于所述雷达传感器在不同时间点的位姿数据、以及在相应时间点采集的三维点云数据中的点相对于所述雷达传感器的距离信息，确定所述拼接点云数据。In a possible implementation manner, based on the pose data of the radar sensor at different time points and the three-dimensional point cloud data collected at the corresponding time points, it is determined that the three-dimensional point cloud data of the target area is obtained after splicing. The splicing point cloud data includes: determining the splicing based on the pose data of the radar sensor at different time points and the distance information of the points in the three-dimensional point cloud data collected at the corresponding time points relative to the radar sensor. point cloud data.

在一种可能的实施方式中，所述基于所述雷达传感器在不同时间点的位姿数据、以及在相应时间点采集的三维点云数据中的点相对于所述雷达传感器的距离信息，确定所述拼接点云数据，包括：基于所述雷达传感器在不同时间点的位姿数据、以及在相应时间点采集的三维点云数据中的点相对于所述雷达传感器的距离信息，确定所述不同时间点下，所述目标区域的三维点云数据中的各个点的位置信息；基于所述不同时间点下，所述目标区域的三维点云数据中的各个点的位置信息，对不同时间点下所述目标区域的三维点云数据中的点进行拼接，得到所述拼接点云数据。In a possible implementation manner, the determination is based on the pose data of the radar sensor at different time points and the distance information of the points in the three-dimensional point cloud data collected at the corresponding time points relative to the radar sensor. The splicing point cloud data includes: based on the pose data of the radar sensor at different time points and the distance information of the points in the three-dimensional point cloud data collected at the corresponding time point relative to the radar sensor, determining the At different time points, the position information of each point in the three-dimensional point cloud data of the target area; based on the position information of each point in the three-dimensional point cloud data of the target area at the different time points, for different times Click the points in the three-dimensional point cloud data of the target area for splicing to obtain the spliced point cloud data.

本公开实施例中，在雷达传感器在不同时间点下的位姿数据为在相同坐标系下的位姿数据时，可以基于雷达传感器的位姿数据和距离信息，确定出不同时间点下的三维点云数据中的点在相同坐标系下的位置信息，基于此，可以对不同时间点下目标区域的三维点云数据中的点进行拼接，得到拼接点云数据，为进行外部参数信息进行优化提供支持。In the embodiment of the present disclosure, when the pose data of the radar sensor at different time points are the pose data in the same coordinate system, the three-dimensional pose data at different time points can be determined based on the pose data and distance information of the radar sensor. The position information of the points in the point cloud data in the same coordinate system. Based on this, the points in the 3D point cloud data of the target area at different time points can be spliced to obtain the spliced point cloud data, which is optimized for external parameter information. provide support.

在一种可能的实施方式中，基于所述不同时间点下所述目标区域对应的三维点云数据中的各个点的位置信息，对不同时间点下所述目标区域对应的三维点云数据中的点进行拼接，得到所述拼接点云数据，包括：基于所述不同时间点下所述目标区域对应的三维点云数据中的各个点的位置信息，对所述目标区域对应的三维点云数据中位置重复的点进行去重处理；基于进行所述去重处理后、所述不同时间点下所述目标区域对应的三维点云数据中的各个点的位置信息，对不同时间点下所述目标区域对应的三维点云数据中的点进行拼接，得到所述拼接点云数据。In a possible implementation manner, based on the position information of each point in the 3D point cloud data corresponding to the target area at the different time points, for the 3D point cloud data corresponding to the target area at different time points splicing the spliced points to obtain the spliced point cloud data, including: based on the position information of each point in the 3D point cloud data corresponding to the target area at the different time points, splicing the 3D point cloud corresponding to the target area The duplicated points in the data are subjected to de-duplication processing; based on the position information of each point in the 3D point cloud data corresponding to the target area at the different time points after the de-duplication processing, The points in the three-dimensional point cloud data corresponding to the target area are spliced to obtain the spliced point cloud data.

本公开实施例中，提出在对目标区域在不同时间点下对应的三维点云数据中的点进行拼接之前，先对位置重复的点进行去重处理，从而减少后期数据处理时的冗余，以提高标定速度。In the embodiment of the present disclosure, it is proposed that before splicing the points in the 3D point cloud data corresponding to the target area at different time points, de-duplication processing is performed on the points with repeated positions, thereby reducing the redundancy in the later data processing. to increase the calibration speed.

在一种可能的实施方式中，所述基于所述拼接点云数据，对所述外部参数信息进行调整，得到目标外部参数信息，包括：根据所述拼接点云数据，对所述外部参数信息进行调整；确定调整后的外部参数信息与调整前的外部参数信息不同，基于调整后的外部参数信息重新确定拼接点云数据后，返回根据所述拼接点云数据，对所述外部参数信息进行调整的过程，直至调整后的外部参数信息与调整前的外部参数信息相同；将调整后的所述外部参数信息作为所述目标外部参数信息。In a possible implementation manner, the adjusting the external parameter information based on the splicing point cloud data to obtain the target external parameter information includes: adjusting the external parameter information according to the splicing point cloud data Adjustment is performed; it is determined that the external parameter information after adjustment is different from the external parameter information before adjustment, after re-determining the splicing point cloud data based on the adjusted external parameter information, and returning to the splicing point cloud data, the external parameter information is processed. The adjustment process is performed until the external parameter information after adjustment is the same as the external parameter information before adjustment; the external parameter information after adjustment is used as the target external parameter information.

本公开实施例中，考虑到在外部参数信息不再准确的情况下，目标区域对应的拼接点云数据中的点的个数会相比一个时间点下的三维点云数据中的点的个数会变多，因此，可以基于拼接点云数据，对外部参数信息进行调整，直至调整后的外部参数信息与调整前的外部参数信息相同时，可以得到准确度较高的目标外部参数信息。In the embodiment of the present disclosure, considering that when the external parameter information is no longer accurate, the number of points in the spliced point cloud data corresponding to the target area will be compared with the number of points in the three-dimensional point cloud data at a time point. Therefore, the external parameter information can be adjusted based on the spliced point cloud data, until the adjusted external parameter information is the same as the pre-adjusted external parameter information, the target external parameter information with higher accuracy can be obtained.

在一种可能的实施方式中，所述外部参数信息包含多个外部参数，所述根据所述拼接点云数据，对所述外部参数信息进行调整，包括：对所述外部参数信息中的至少一个外部参数进行调整，得到调整后的外部参数信息。In a possible implementation manner, the external parameter information includes multiple external parameters, and the adjusting the external parameter information according to the spliced point cloud data includes: adjusting at least one of the external parameter information An external parameter is adjusted to obtain the adjusted external parameter information.

本公开实施例中，保护在外部参数信息包含多个外部参数的情况下，可以基于至少一个外部参数对外部参数信息进行细微的调整，从而得到准确度较高的目标外部参数信息。In the embodiment of the present disclosure, when the external parameter information includes multiple external parameters, the external parameter information may be finely adjusted based on at least one external parameter, thereby obtaining target external parameter information with higher accuracy.

在一种可能的实施方式中，所述对所述外部参数信息中的至少一个外部参数进行调整，得到调整后的外部参数信息，包括：在对所述外部参数信息进行当前轮调整过程中，从所述多个外部参数中选择当前轮还未调整的外部参数；其中，对外部参数信息进行一轮调整包括对所述外部参数信息中的各个外部参数均进行调整；对选择的外部参数的参数值进行当前次调整，得到当前调整后的外部参数信息；确定基于当前调整后的外部参数信息得到的拼接点云数据中的点的个数与基于当前次调整前的外部参数信息得到的拼接点云数据中的点的个数相比是否变少；其中，当前次调整前的外部参数信息包括选择的外部参数在进行当前次调整之前的参数值；若变少，对选择的外部参数的参数值进行更新后，返回对选择的外部参数进行当前次调整的过程，否则，保持选择的外部参数在当前次调整前的参数值，并返回从所述多个外部参数中选择当前轮还未调整的外部参数的过程；在对所述外部参数信息完成当前轮调整后，判断当前轮最后选择的外部参数的参数值在当前轮调整前后是否发生变化；若发生变化，进行下一轮外部参数信息调整，若未发生变化，确定对外部参数信息的调整结果达到调整后的外部参数信息与调整前的外部参数信息相同的条件，将调整后的外部参数信息作为所述目标外部参数信息。In a possible implementation manner, the adjusting at least one external parameter in the external parameter information to obtain the adjusted external parameter information includes: during the current round of adjusting the external parameter information, Selecting an external parameter that has not been adjusted in the current round from the plurality of external parameters; wherein, performing one round of adjustment on the external parameter information includes adjusting each external parameter in the external parameter information; The parameter value is adjusted for the current time, and the external parameter information after the current adjustment is obtained; the number of points in the splicing point cloud data obtained based on the external parameter information after the current adjustment is determined and the splicing obtained based on the external parameter information before the current adjustment is determined. Whether the number of points in the point cloud data is reduced compared to that; wherein, the external parameter information before the current adjustment includes the parameter value of the selected external parameter before the current adjustment; After the parameter value is updated, return to the process of the current adjustment of the selected external parameter, otherwise, keep the parameter value of the selected external parameter before the current adjustment, and return to select the current round from the multiple external parameters. The process of adjusting the external parameters; after completing the current round of adjustment for the external parameter information, determine whether the parameter value of the external parameter last selected by the current round has changed before and after the current round of adjustment; if there is a change, carry out the next round of external parameters In the information adjustment, if there is no change, it is determined that the adjustment result of the external parameter information reaches the same condition as the external parameter information after adjustment and the external parameter information before adjustment, and the adjusted external parameter information is used as the target external parameter information.

本公开实施例中，在优化外部参数信息的过程中，可以基于拼接点云数据中点的个数的变化，对外部参数信息进行不断优化，直至使得到准确度较高的外部参数信息。In the embodiment of the present disclosure, in the process of optimizing the external parameter information, the external parameter information can be continuously optimized based on the change of the number of points in the spliced point cloud data until the external parameter information with higher accuracy is obtained.

在一种可能的实施方式中，在得到所述目标外部参数信息后，所述标定方法还包括：In a possible implementation manner, after obtaining the target external parameter information, the calibration method further includes:

获取所述行驶设备行驶过程中，所述组合惯性导航设备的位姿数据、以及所述雷达传感器采集的三维点云数据；基于所述目标外部参数信息和所述组合惯性导航设备的位姿数据，确定所述雷达传感器的位姿数据；基于所述雷达传感器在不同时间点的位姿数据、以及在相应时间点采集的三维点云数据，确定所述三维点云数据中各个点的位置信息；基于所述三维点云数据中各个点的位置信息，构建所述行驶设备行驶的区域的地图。Acquire the pose data of the integrated inertial navigation device and the three-dimensional point cloud data collected by the radar sensor during the driving process of the traveling device; based on the target external parameter information and the pose data of the integrated inertial navigation device , determine the pose data of the radar sensor; based on the pose data of the radar sensor at different time points and the three-dimensional point cloud data collected at the corresponding time point, determine the position information of each point in the three-dimensional point cloud data ; Based on the position information of each point in the three-dimensional point cloud data, a map of the area where the traveling device travels is constructed.

本公开实施例中，在得到准确度较高的目标外部参数信息后，可以基于组合惯性导航设备和雷达传感器结合的方式准确地得到行驶车辆行驶的区域对应的三维点云数据中各个点的位置信息，从而便于针对该行驶车辆行驶的区域构建准确度较高的地图。In the embodiment of the present disclosure, after obtaining the target external parameter information with high accuracy, the position of each point in the three-dimensional point cloud data corresponding to the area where the driving vehicle is traveling can be accurately obtained based on the combination of the combined inertial navigation device and the radar sensor. information, so as to facilitate the construction of a map with higher accuracy for the area where the traveling vehicle travels.

第二方面，本公开实施例提供了一种标定装置，包括：获取模块，用于获取在行驶设备行驶过程中，所述行驶设备上的组合惯性导航设备的位姿数据、以及所述行驶设备上的雷达传感器采集的三维点云数据；确定模块，用于基于表征所述雷达传感器与所述组合惯性导航设备之间坐标系转换关系的外部参数信息、以及所述组合惯性导航设备的位姿数据，确定所述雷达传感器的位姿数据；拼接模块，用于基于所述雷达传感器在不同时间点的位姿数据、以及在相应时间点采集的三维点云数据，确定对目标区域的三维点云数据进行拼接后得到的拼接点云数据；调整模块，用于基于所述拼接点云数据，对所述外部参数信息进行调整，得到目标外部参数信息。In a second aspect, an embodiment of the present disclosure provides a calibration apparatus, including: an acquisition module configured to acquire pose data of an integrated inertial navigation device on the traveling device during the traveling process of the traveling device, and the traveling device The three-dimensional point cloud data collected by the radar sensor on the device; the determination module is used for the external parameter information representing the coordinate system transformation relationship between the radar sensor and the integrated inertial navigation device, and the pose of the integrated inertial navigation device. data, to determine the pose data of the radar sensor; the splicing module is used to determine the three-dimensional point of the target area based on the pose data of the radar sensor at different time points and the three-dimensional point cloud data collected at the corresponding time point The spliced point cloud data obtained after the cloud data is spliced; the adjustment module is used to adjust the external parameter information based on the spliced point cloud data to obtain target external parameter information.

在一种可能的实施方式中，所述拼接模块用于：基于所述雷达传感器在不同时间点的位姿数据、以及在相应时间点采集的三维点云数据中的点相对于所述雷达传感器的距离信息，确定所述拼接点云数据。In a possible implementation manner, the splicing module is configured to: relative to the radar sensor based on the pose data of the radar sensor at different time points and the points in the three-dimensional point cloud data collected at the corresponding time point distance information to determine the spliced point cloud data.

在一种可能的实施方式中，所述拼接模块用于：基于所述雷达传感器在不同时间点的位姿数据、以及在相应时间点采集的三维点云数据中的点相对于所述雷达传感器的距离信息，确定所述不同时间点下，所述目标区域的三维点云数据中的各个点的位置信息；基于所述不同时间点下，所述目标区域的三维点云数据中的各个点的位置信息，对不同时间点下所述目标区域的三维点云数据中的点进行拼接，得到所述拼接点云数据。In a possible implementation manner, the splicing module is configured to: relative to the radar sensor based on the pose data of the radar sensor at different time points and the points in the three-dimensional point cloud data collected at the corresponding time point distance information, determine the position information of each point in the 3D point cloud data of the target area at the different time points; based on the different time points, each point in the 3D point cloud data of the target area The location information is obtained by splicing the points in the three-dimensional point cloud data of the target area at different time points to obtain the spliced point cloud data.

在一种可能的实施方式中，所述拼接模块在基于所述不同时间点下所述目标区域对应的三维点云数据中的各个点的位置信息，在对不同时间点下所述目标区域对应的三维点云数据中的点进行拼接，得到所述拼接点云数据时，用于：基于所述不同时间点下所述目标区域对应的三维点云数据中的各个点的位置信息，对所述目标区域对应的三维点云数据中位置重复的点进行去重处理；基于进行所述去重处理后、所述不同时间点下所述目标区域对应的三维点云数据中的各个点的位置信息，对不同时间点下所述目标区域对应的三维点云数据中的点进行拼接，得到所述拼接点云数据。In a possible implementation manner, the splicing module is based on the position information of each point in the three-dimensional point cloud data corresponding to the target area at the different time points, and the target area corresponds to the target area at different time points. When the points in the three-dimensional point cloud data are spliced, and the spliced point cloud data is obtained, it is used for: based on the position information of each point in the three-dimensional point cloud data corresponding to the target area at the different time points, Perform de-duplication processing on the duplicated points in the three-dimensional point cloud data corresponding to the target area; based on the position of each point in the three-dimensional point cloud data corresponding to the target area at the different time points after the de-duplication processing is performed information, and splicing points in the three-dimensional point cloud data corresponding to the target area at different time points to obtain the spliced point cloud data.

在一种实施方式中，所述调整模块用于：根据所述拼接点云数据，对所述外部参数信息进行调整；确定调整后的外部参数信息与调整前的外部参数信息不同，基于调整后的外部参数信息重新确定拼接点云数据后，返回根据所述拼接点云数据，对所述外部参数信息进行调整的过程，直至调整后的外部参数信息与调整前的外部参数信息相同；将调整后的外部参数信息作为所述目标外部参数信息。In one embodiment, the adjustment module is configured to: adjust the external parameter information according to the spliced point cloud data; determine that the external parameter information after adjustment is different from the external parameter information before adjustment, and based on the adjusted external parameter information After re-determining the splicing point cloud data according to the splicing point cloud data, return to the process of adjusting the external parameter information according to the splicing point cloud data, until the adjusted external parameter information is the same as the external parameter information before the adjustment; The subsequent external parameter information is used as the target external parameter information.

在一种实施方式中，所述外部参数信息包含多个外部参数，所述调整模块在用于根据所述拼接点云数据，对所述外部参数信息进行调整时，包括：对所述外部参数信息中的至少一个外部参数进行调整，得到调整后的外部参数信息。In an embodiment, the external parameter information includes multiple external parameters, and when the adjustment module is configured to adjust the external parameter information according to the spliced point cloud data, the method includes: adjusting the external parameter At least one external parameter in the information is adjusted to obtain the adjusted external parameter information.

在一种实施方式中，所述调整模块在用于对所述外部参数信息中的至少一个外部参数进行调整，得到调整后的外部参数信息时，包括：在对所述外部参数信息进行当前轮调整过程中，从所述多个外部参数中选择当前轮还未调整的外部参数；其中，对外部参数信息进行一轮调整包括对所述外部参数信息中的各个外部参数均进行调整；对选择的外部参数的参数值进行当前次调整，得到当前调整后的外部参数信息；确定基于当前调整后的外部参数信息得到的拼接点云数据中的点的个数与基于当前次调整前的外部参数信息得到的拼接点云数据中的点的个数相比是否变少；其中，当前次调整前的外部参数信息包括选择的外部参数在进行当前次调整之前的参数值；若变少，对选择的外部参数的参数值进行更新后，返回对选择的外部参数进行当前次调整的过程，否则，保持选择的外部参数在当前次调整前的参数值，返回从所述多个外部参数中选择当前轮还未调整的外部参数的过程；在对所述外部参数信息完成当前轮调整后，判断当前轮最后选择的外部参数的参数值在当前轮调整前后是否发生变化；若发生变化，进行下一轮外部参数信息调整，若未发生变化，确定对外部参数信息的调整结果达到调整后的外部参数信息与调整前的外部参数信息相同的条件，将调整后的外部参数信息作为所述目标外部参数信息。In an implementation manner, when the adjustment module is configured to adjust at least one external parameter in the external parameter information to obtain the adjusted external parameter information, the adjusting module includes: performing a current round of the external parameter information on the external parameter information. During the adjustment process, the external parameters that have not been adjusted in the current round are selected from the plurality of external parameters; wherein, performing one round of adjustment on the external parameter information includes adjusting each external parameter in the external parameter information; The parameter value of the external parameter is adjusted for the current time to obtain the external parameter information after the current adjustment; determine the number of points in the splicing point cloud data obtained based on the external parameter information after the current adjustment and the external parameter Whether the number of points in the spliced point cloud data obtained from the information is reduced compared to that; wherein, the external parameter information before the current adjustment includes the parameter value of the selected external parameter before the current adjustment; After updating the parameter value of the external parameter of the The process of the external parameters that have not been adjusted in the round; after completing the adjustment of the current round of the external parameter information, determine whether the parameter value of the external parameter finally selected in the current round has changed before and after the adjustment of the current round; If there is no change, it is determined that the adjustment result of the external parameter information reaches the same condition as the external parameter information after adjustment and the external parameter information before adjustment, and the external parameter information after adjustment is used as the target external parameter information.

在一种可能的实施方式中，所述标定装置还包括构图模块，在得到所述目标外部参数信息后，所述构图模块用于：获取所述行驶设备行驶过程中，所述组合惯性导航设备的位姿数据、以及所述雷达传感器采集的三维点云数据；基于所述目标外部参数信息和所述组合惯性导航设备的位姿数据，确定所述雷达传感器的位姿数据；基于所述雷达传感器在不同时间点的位姿数据、以及在相应时间点采集的三维点云数据，确定在所述三维点云数据中各个点的位置信息；基于所述三维点云数据中各个点的位置信息，构建所述行驶设备行驶的区域的地图。In a possible implementation manner, the calibration device further includes a composition module, and after obtaining the target external parameter information, the composition module is configured to: The pose data of the radar sensor and the three-dimensional point cloud data collected by the radar sensor; based on the target external parameter information and the pose data of the combined inertial navigation device, determine the pose data of the radar sensor; based on the radar sensor The position and attitude data of the sensor at different time points and the three-dimensional point cloud data collected at the corresponding time points are used to determine the position information of each point in the three-dimensional point cloud data; based on the position information of each point in the three-dimensional point cloud data , build a map of the area where the traveling device travels.

第三方面，本公开实施例提供了一种电子设备，包括处理器、存储器和总线。所述存储器存储有所述处理器可执行的机器可读指令。当电子设备运行时，所述处理器与所述存储器之间通过总线通信，所述机器可读指令被所述处理器执行时执行如第一方面所述的标定方法。In a third aspect, embodiments of the present disclosure provide an electronic device including a processor, a memory, and a bus. The memory stores machine-readable instructions executable by the processor. When the electronic device is running, the processor and the memory communicate through a bus, and when the machine-readable instructions are executed by the processor, the calibration method according to the first aspect is performed.

第四方面，本公开实施例提供了一种计算机可读存储介质，其上存储有计算机程序，该计算机程序被处理器运行时执行如第一方面所述的标定方法。In a fourth aspect, an embodiment of the present disclosure provides a computer-readable storage medium on which a computer program is stored, and when the computer program is run by a processor, the calibration method according to the first aspect is executed.

为使本公开的上述目的、特征和优点能更明显易懂，下文特举较佳实施例，并配合所附附图，作详细说明如下。In order to make the above-mentioned objects, features and advantages of the present disclosure more obvious and easy to understand, the preferred embodiments are exemplified below, and are described in detail as follows in conjunction with the accompanying drawings.

附图说明Description of drawings

为了更清楚地说明本公开实施例的技术方案，下面将对实施例中所需要使用的附图作简单地介绍。这些附图示出了符合本公开的实施例，并与说明书一起用于说明本公开的技术方案。应当理解，以下附图仅示出了本公开的某些实施例，因此不应被看作是对范围的限定，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他相关的附图。In order to illustrate the technical solutions of the embodiments of the present disclosure more clearly, the accompanying drawings required in the embodiments will be briefly introduced below. These drawings illustrate embodiments consistent with the present disclosure, and together with the description, serve to explain the technical solutions of the present disclosure. It should be understood that the following drawings only show some embodiments of the present disclosure, and therefore should not be regarded as limiting the scope. Other related figures are obtained from these figures.

图1示出了本公开实施例所提供的一种标定方法的流程图；1 shows a flowchart of a calibration method provided by an embodiment of the present disclosure;

图2示出了本公开实施例所提供的一种针对外部参数信息进行调整的方法流程图；FIG. 2 shows a flowchart of a method for adjusting external parameter information provided by an embodiment of the present disclosure;

图3示出了本公开实施例所提供的一种针对外部参数信息中的至少一个外部参数进行调整的方法流程图；3 shows a flowchart of a method for adjusting at least one external parameter in external parameter information provided by an embodiment of the present disclosure;

图4示出了本公开实施例所提供的一种标定装置的结构示意图；FIG. 4 shows a schematic structural diagram of a calibration device provided by an embodiment of the present disclosure;

图5示出了本公开实施例所提供的一种电子设备的结构示意图。FIG. 5 shows a schematic structural diagram of an electronic device provided by an embodiment of the present disclosure.

具体实施方式Detailed ways

为使本公开实施例的目的、技术方案和优点更加清楚，下面将结合本公开实施例中附图，对本公开实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例仅仅是本公开一部分实施例，而不是全部的实施例。通常在此处附图中描述和示出的本公开实施例的组件可以以各种不同的配置来布置和设计。因此，以下对在附图中提供的本公开的实施例的详细描述并非旨在限制要求保护的本公开的范围，而是仅仅表示本公开的选定实施例。基于本公开的实施例，本领域技术人员在没有做出创造性劳动的前提下所获得的所有其他实施例，都属于本公开保护的范围。In order to make the purposes, technical solutions and advantages of the embodiments of the present disclosure more clear, the technical solutions in the embodiments of the present disclosure will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present disclosure. Obviously, the described embodiments are only These are some, but not all, embodiments of the present disclosure. The components of the disclosed embodiments generally described and illustrated in the drawings herein may be arranged and designed in a variety of different configurations. Therefore, the following detailed description of the embodiments of the disclosure provided in the accompanying drawings is not intended to limit the scope of the disclosure as claimed, but is merely representative of selected embodiments of the disclosure. Based on the embodiments of the present disclosure, all other embodiments obtained by those skilled in the art without creative work fall within the protection scope of the present disclosure.

应注意到：相似的标号和字母在下面的附图中表示类似项，因此，一旦某一项在一个附图中被定义，则在随后的附图中不需要对其进行进一步定义和解释。It should be noted that like numerals and letters refer to like items in the following figures, so once an item is defined in one figure, it does not require further definition and explanation in subsequent figures.

在行驶设备行驶过程中，可以通过设置于行驶设备上的不同定位传感器来对行驶设备进行定位，比如通过设置于行驶设备上的组合惯性导航设备以及雷达传感器来进行综合定位。但是不同传感器检测的位姿数据所属的坐标系不同，在基于组合惯性导航设备和雷达传感器进行定位时，需要将组合惯性导航设备的位姿数据和雷达传感器的位姿数据通过外部参数信息转换至相同的坐标系下。这样，外部参数信息标定的准确性将直接影响定位的准确性。During the traveling process of the traveling device, the traveling device can be positioned by different positioning sensors provided on the traveling device, for example, the integrated inertial navigation device and the radar sensor provided on the traveling device can perform comprehensive positioning. However, the coordinate systems of the pose data detected by different sensors are different. When positioning based on the combined inertial navigation device and the radar sensor, it is necessary to convert the pose data of the combined inertial navigation device and the radar sensor to the external parameter information. in the same coordinate system. In this way, the accuracy of external parameter information calibration will directly affect the accuracy of positioning.

可以采用基于手眼标定方式来对组合惯性导航设备和雷达传感器之间的外部参数信息进行标定，具体过程可以是：使得配置有组合惯性导航设备和雷达传感器的车辆按照预设路线进行行驶后，获取组合惯性导航设备在不同时间点在世界坐标系下的位姿数据、以及获取在对应时间点基于不同目标物与雷达传感器的距离、确定的雷达传感器在雷达坐标系下的位姿数据；在得到不同时间点下的组合惯性导航设备在世界坐标系下的位姿数据、以及雷达传感器在雷达坐标下的位姿数据后，根据手眼标定算法确定出组合惯性导航设备和雷达传感器之间的外部参数信息。通过该方式，由于在对应时间点基于不同目标物与雷达传感器的距离确定的雷达传感器在雷达坐标系下的位姿数据的精度较低，因此最终确定的外部参数信息的精度不高。The external parameter information between the combined inertial navigation device and the radar sensor can be calibrated by a hand-eye-based calibration method. The specific process can be as follows: after the vehicle equipped with the combined inertial navigation device and the radar sensor is driven according to the preset route, obtain Combine the position and attitude data of the inertial navigation device in the world coordinate system at different time points, and obtain the position and attitude data of the radar sensor in the radar coordinate system based on the distance between different targets and the radar sensor at the corresponding time point; After the pose data of the combined inertial navigation device in the world coordinate system and the pose data of the radar sensor in the radar coordinates at different time points, the external parameters between the combined inertial navigation device and the radar sensor are determined according to the hand-eye calibration algorithm information. In this way, the accuracy of the pose data of the radar sensor in the radar coordinate system determined based on the distances between different targets and the radar sensor at the corresponding time point is low, so the accuracy of the finally determined external parameter information is not high.

基于上述研究，本公开发明人提出：在通过预先确定的外部参数信息确定出雷达传感器在不同时间点的位姿数据后，可以结合雷达传感器在不同时间点采集的三维点云数据确定出不同时间点下表征目标区域的三维点云数据；这样，在外部参数信息不再准确的情况下，不同时间点下表征目标区域同一位置点的坐标值不再相同，并使得拼接点云数据中的点变多；基于此，可以基于拼接点云数据对外部参数信息进行调整，从而不断优化外部参数信息，直至得到准确度较高的目标外部参数信息。Based on the above research, the inventors of the present disclosure propose that: after determining the pose data of the radar sensor at different time points through the predetermined external parameter information, the three-dimensional point cloud data collected by the radar sensor at different time points can be combined to determine the different time points. The three-dimensional point cloud data representing the target area under the point; in this way, when the external parameter information is no longer accurate, the coordinate values of the same point representing the target area at different time points are no longer the same, and the points in the point cloud data are spliced. Based on this, the external parameter information can be adjusted based on the spliced point cloud data, so as to continuously optimize the external parameter information until the target external parameter information with higher accuracy is obtained.

为便于对本实施例进行理解，首先对本公开实施例所公开的一种标定方法进行详细介绍，本公开实施例所提供的标定方法的执行主体一般为具有一定计算能力的计算机设备，该计算机设备例如包括终端设备或服务器或其它处理设备，终端设备可以为用户设备(User Equipment，UE)、移动设备、手持设备、计算设备或车载设备等。在一些可能的实现方式中，该标定方法可以通过处理器调用存储器中存储的计算机可读指令的方式来实现。In order to facilitate the understanding of this embodiment, a calibration method disclosed by the embodiment of the present disclosure is first introduced in detail. The execution subject of the calibration method provided by the embodiment of the present disclosure is generally a computer device with a certain computing capability, such as a computer device. Including a terminal device or a server or other processing device, the terminal device may be a user equipment (User Equipment, UE), a mobile device, a handheld device, a computing device, or a vehicle-mounted device. In some possible implementations, the calibration method may be implemented by the processor invoking computer-readable instructions stored in the memory.

参见图1所示，为本公开实施例提供的一种标定方法的流程图，该标定方法包括步骤S101～S104：Referring to FIG. 1, which is a flowchart of a calibration method provided by an embodiment of the present disclosure, the calibration method includes steps S101-S104:

S101，获取在行驶设备行驶过程中，行驶设备上的组合惯性导航设备的位姿数据、以及行驶设备上的雷达传感器采集的三维点云数据。S101: Acquire pose data of an integrated inertial navigation device on the traveling device and three-dimensional point cloud data collected by a radar sensor on the traveling device during the traveling process of the traveling device.

示例性地，行驶设备可以为安装有组合惯性导航设备和雷达传感器的能够行驶的设备，例如包括行驶的车辆、行驶的机器人等，本公开实施例中以行驶设备为行驶车辆为例进行阐述。Exemplarily, the traveling device may be a traveling device equipped with a combined inertial navigation device and a radar sensor, for example, including a traveling vehicle, a traveling robot, and the like.

示例性地，组合惯性导航设备可以是惯性测量单元(Inertial measurement unit，IMU)和全球定位***(Global Positioning System，GPS)组成的组合惯性导航设备，在行驶设备行驶过程中，其能够输出该组合惯性导航设备的位姿数据。当该组合惯性导航设备设置于行驶设备上时，组合惯性导航设备输出的位姿数据也可以表示行驶设备的位姿数据。Exemplarily, the combined inertial navigation device may be a combined inertial navigation device composed of an inertial measurement unit (Inertial measurement unit, IMU) and a global positioning system (Global Positioning System, GPS), and during the driving process of the traveling device, it can output the combined inertial navigation device. Pose data of inertial navigation equipment. When the integrated inertial navigation device is installed on the traveling device, the pose data output by the integrated inertial navigation device can also represent the pose data of the traveling device.

示例性地，组合惯性导航设备的位姿数据是指组合惯性导航设备在世界坐标系中的位姿数据，其中，这里的世界坐标系可以包括球面坐标系和直角坐标系(笛卡尔坐标系)。其中组合惯性导航设备在球面坐标系中的坐标是指该组合惯性导航设置的经度、纬度、高度以及朝向。组合惯性导航设置在直角坐标系中的坐标是指组合惯性导航设置分别沿该直角坐标系的X轴、Y轴和Z轴方向平移的距离，以及分别沿X轴、Y轴和Z轴方向旋转的角度，可以通过(X，Y，Z，Roll，Pitch，Yaw)进行表示。为了方便后期运算，这里得到的组合惯性导航设备在世界坐标系下的位姿数据是指组合惯性导航设备在直角坐标系下的位姿数据。Exemplarily, the pose data of the combined inertial navigation device refers to the pose data of the combined inertial navigation device in the world coordinate system, wherein the world coordinate system here may include a spherical coordinate system and a Cartesian coordinate system (Cartesian coordinate system) . The coordinates of the integrated inertial navigation device in the spherical coordinate system refer to the longitude, latitude, altitude and orientation set by the integrated inertial navigation. The coordinates of the combined inertial navigation setting in the Cartesian coordinate system refer to the distance that the combined inertial navigation setting translates along the X-axis, Y-axis and Z-axis of the Cartesian coordinate system, and rotates along the X-axis, Y-axis and Z-axis respectively. The angle can be represented by (X, Y, Z, Roll, Pitch, Yaw). In order to facilitate later operations, the pose data of the combined inertial navigation device in the world coordinate system obtained here refers to the pose data of the combined inertial navigation device in the rectangular coordinate system.

具体地，可以以行驶设备出发地所在的位置点为坐标原点，以从该位置点开始，行驶设备前进方向为X轴，以垂直于行驶设备前进方向且水平向左的方向为Y轴，以垂直于行驶设备前进方向且指向天空的方向为Z轴，来建立世界坐标系。将组合惯性导航设备放置于行驶设备的设定位置时，即可以通过组合惯性导航设备在各个第一时间点采集的该组合惯性导航设备在球面坐标系下的位姿数据以及球面坐标系和直接坐标系之间的转换关系，得到该组合惯性导航设备在直角坐标系下的位姿数据。Specifically, the position of the starting point of the traveling equipment can be taken as the origin of the coordinates, starting from this point, the forward direction of the traveling equipment is the X axis, and the direction perpendicular to the traveling direction of the traveling equipment and horizontally to the left is the Y axis, and The direction perpendicular to the forward direction of the traveling device and pointing to the sky is the Z axis to establish the world coordinate system. When the combined inertial navigation device is placed at the set position of the driving device, the pose data of the combined inertial navigation device in the spherical coordinate system, the spherical coordinate system and the direct The transformation relationship between the coordinate systems is used to obtain the pose data of the combined inertial navigation device in the Cartesian coordinate system.

示例性地，行驶设备上的雷达传感器采集的三维点云数据是指在雷达传感器对应的雷达坐标系下，雷达传感器采集到的目标区域中包含的位置点相对于雷达传感器的距离信息。具体在行驶设备行驶过程中，可以获取到行驶设备上的雷达传感器按照设定传输时间间隔传输的三维点云数据，基于获取到的雷达传感器传输的三维点云数据，可以得到雷达传感器按照设定采集时间间隔采集的三维点云数据。具体地，雷达传感传感器传输三维点云数据的传输时间间隔和雷达传感器采集三维点云数据的采集时间间隔可能不同，比如每隔1s采集一次三维点云数据，然后每隔2s传输一次采集到的三维点云数据。Exemplarily, the three-dimensional point cloud data collected by the radar sensor on the traveling device refers to the distance information of the position points included in the target area collected by the radar sensor relative to the radar sensor in the radar coordinate system corresponding to the radar sensor. Specifically, during the driving process of the driving equipment, the 3D point cloud data transmitted by the radar sensor on the driving equipment according to the set transmission time interval can be obtained. Based on the acquired 3D point cloud data transmitted by the radar sensor, the radar sensor can be obtained according to the set Collect 3D point cloud data collected at time intervals. Specifically, the transmission time interval for the radar sensor to transmit 3D point cloud data may be different from the collection time interval for the radar sensor to collect 3D point cloud data. For example, the 3D point cloud data is collected every 1s, and then the collected data is transmitted every 2s. 3D point cloud data.

示例性地，这里可以获取行驶设备上的组合惯性导航设备在各个第一时间点的位姿数据、以及雷达传感器在各个第二时间点采集的三维点云数据。考虑到雷达传感器采集三维点云数据的采集时间间隔和传输三维点云数据的传输时间间隔可能不同，因此这里获取到的雷达传感器在各个第二时间点采集的三维点云数据可以是针对获取到的雷达传感器在各个第二时间点传输的三维点云数据进行去畸变处理后得到的，去畸变处理的过程具体将在后文进行解释说明。Exemplarily, the pose data of the integrated inertial navigation device on the traveling device at each first time point and the three-dimensional point cloud data collected by the radar sensor at each second time point can be acquired here. Considering that the collection time interval for the radar sensor to collect the 3D point cloud data and the transmission time interval for the transmission of the 3D point cloud data may be different, the 3D point cloud data collected by the radar sensor at each second time point can be obtained for the acquired 3D point cloud data. The 3D point cloud data transmitted by the radar sensor at each second time point is obtained after de-distortion processing, and the process of de-distortion processing will be explained in detail later.

示例性地，第一时间点是指从设定时刻开始，比如从行驶设备开始行驶的时刻开始，组合惯性导航设备按照设定采集时间间隔进行采集时对应的时间点可以称为这里的第一时间点，比如行驶设备开始行驶的时刻为t，若组合惯性导航设备是按照每隔Δt1秒采集一次位姿数据，则这里的第一时间点包括t+Δt1，t+2Δt1，t+3Δt1，…。这里的第二时间点是指从设定时刻开始，比如同样从行驶设备开始行驶的时刻开始，雷达传感器按照设定传输时间间隔传输三维点云数据时对应的时间点即为这里的第二时间点，比如行驶设备开始行驶的时刻为t，若雷达传感器是按照每隔Δt2秒传输一次三维点云数据，则这里的第二时间点包括t+Δt2，t+2Δt2，t+3Δt2，…。Exemplarily, the first time point refers to starting from a set time, for example, starting from the time when the traveling device starts to drive, and the time corresponding to the time when the integrated inertial navigation device collects according to the set collection time interval may be referred to as the first time here. Time point, for example, the time when the driving equipment starts to drive is t. If the combined inertial navigation equipment collects pose data every Δt1 seconds, the first time point here includes t+Δt1, t+2Δt1, t+3Δt1, …. The second time point here refers to starting from the set time, for example, starting from the time when the driving equipment starts to drive, and the corresponding time point when the radar sensor transmits the 3D point cloud data according to the set transmission time interval is the second time here. For example, the time when the driving equipment starts to drive is t. If the radar sensor transmits three-dimensional point cloud data every Δt2 seconds, the second time point here includes t+Δt2, t+2Δt2, t+3Δt2, . . .

S102，基于表征雷达传感器与组合惯性导航设备之间坐标系转换关系的外部参数信息、以及组合惯性导航设备的位姿数据，确定雷达传感器在的位姿数据。S102 , based on the external parameter information representing the coordinate system transformation relationship between the radar sensor and the integrated inertial navigation device, and the pose data of the integrated inertial navigation device, determine the pose data of the radar sensor.

在对外部参数信息进行调整之前，初始的外部参数信息可以是手工测量得到的，该外部参数信息由6个参数构成，可以通过(X，Y，Z，Roll，Pitch，Yaw)表示，初始确定过程具体如下：Before adjusting the external parameter information, the initial external parameter information can be obtained by manual measurement. The external parameter information is composed of 6 parameters, which can be represented by (X, Y, Z, Roll, Pitch, Yaw). The process is as follows:

比如在建立好世界坐标系后，针对相同的位置点A，通过组合惯性导航设备测量得到的到达该位置点A在世界坐标系下的位姿数据表示为M1，当组合惯性导航设备位于行驶设备上时，该位姿数据M1可以表示该行驶设备的位姿数据。通过雷达传感器测量得到的该位置点A在雷达坐标系下的位姿数据为M2，同样，当该雷达传感器位于行驶设备上时，该位姿数据M2可以表示该行驶设备的位姿数据。当组合惯性导航设备和雷达传感器位于相同的行驶设备上时，组合惯性导航设备测量的位姿数据M1和雷达传感器测量的位姿数据M2可以通过外部参数信息进行转换，这样可以通过位姿数据M1和位姿数据M2，确定该外部参数信息，该外部参数信息可以表征雷达传感器对应的雷达坐标系与组合惯性导航设备对应的世界坐标系之间的坐标系转换关系。For example, after the world coordinate system is established, for the same position point A, the pose data obtained by the combined inertial navigation device to reach the position point A in the world coordinate system is expressed as M1. When the combined inertial navigation device is located in the driving device When on, the pose data M1 may represent the pose data of the traveling device. The pose data of the position point A measured by the radar sensor in the radar coordinate system is M2. Similarly, when the radar sensor is located on the traveling device, the pose data M2 can represent the pose data of the traveling device. When the combined inertial navigation device and the radar sensor are located on the same driving device, the pose data M1 measured by the combined inertial navigation device and the pose data M2 measured by the radar sensor can be converted by external parameter information, so that the pose data M1 can be converted by the external parameter information. and pose data M2 to determine the external parameter information, which can represent the coordinate system conversion relationship between the radar coordinate system corresponding to the radar sensor and the world coordinate system corresponding to the integrated inertial navigation device.

按照以上方式得到初始的外部参数信息后，基于上述外部参数信息、以及组合惯性导航设备在各个第一时间点的位姿数据(在此是指组合惯性导航设备在各个第一时间点时在世界坐标系下的位姿数据)，可以确定雷达传感器在各个第一时间点的位姿数据(在此是指雷达传感器在各个第一时间点在世界坐标系下的位姿数据)，然后可以基于插值法确定雷达传感器在各个第二时间点时在世界坐标系下的位姿数据，该插值过程具体将在后文进行阐述。After the initial external parameter information is obtained in the above manner, based on the above external parameter information and the pose data of the combined inertial navigation device at each first time point (here, the combined inertial navigation device is in the world at each first time point) The pose data in the coordinate system), the pose data of the radar sensor at each first time point (here refers to the pose data of the radar sensor in the world coordinate system at each first time point) can be determined, and then based on The interpolation method determines the pose data of the radar sensor in the world coordinate system at each second time point, and the interpolation process will be described in detail later.

S103，基于雷达传感器在不同时间点的位姿数据、以及在相应时间点采集的三维点云数据，确定对目标区域的三维点云数据进行拼接后得到的拼接点云数据。S103 , based on the pose data of the radar sensor at different time points and the three-dimensional point cloud data collected at the corresponding time points, determine the spliced point cloud data obtained by splicing the three-dimensional point cloud data of the target area.

示例性地，雷达传感器在不同时间点的位姿数据是指上述提到的雷达传感器在各个第二时间点的位姿数据，同样，雷达传感器在相应时间点采集的三维点云数据是指雷达传感器在各个第二时间点采集的三维点云数据。Exemplarily, the pose data of the radar sensor at different time points refers to the pose data of the aforementioned radar sensor at each second time point. Similarly, the three-dimensional point cloud data collected by the radar sensor at the corresponding time point refers to the radar sensor. The three-dimensional point cloud data collected by the sensor at each second time point.

具体地，雷达传感器在各个第二时间点的位姿数据是指在世界坐标系下的位姿数据，然后结合雷达传感器在各个第二时间点采集的三维点云数据，即三维点云数据中的点在雷达坐标系下相对于雷达传感器的距离信息，可以对目标区域对应的三维点云数据中的点进行位置拼接，得到目标区域对应的拼接点云数据。Specifically, the pose data of the radar sensor at each second time point refers to the pose data in the world coordinate system, and then combined with the three-dimensional point cloud data collected by the radar sensor at each second time point, that is, in the three-dimensional point cloud data The distance information of the point relative to the radar sensor in the radar coordinate system can be used to stitch the points in the 3D point cloud data corresponding to the target area to obtain the stitched point cloud data corresponding to the target area.

示例性地，这里对目标区域对应的三维点云数据中的点进行位置拼接，是指不同时刻下，基于三维点云数据中的点在世界坐标系中的位置信息进行拼接，可以得到目标区域对应的拼接点云数据，该过程将在后文进行详细阐述。Exemplarily, the location splicing of the points in the 3D point cloud data corresponding to the target area refers to splicing based on the location information of the points in the 3D point cloud data in the world coordinate system at different times, and the target area can be obtained. The corresponding splicing point cloud data, the process will be described in detail later.

示例性地，目标区域是指同一位置区域，在提前设置好世界坐标系后，该目标位置区域中的各个位置点在该世界坐标系中的坐标值并不会随着行驶设备的行驶而发生变化，因此，在不同时间点采集的三维点云数据中用于表示目标区域相同位置点的点，在外部参数信息准确的情况下在三维点云数据进行拼接后会发生重叠，反之，在外部参数信息不准确的情况下，拼接后得到的拼接点云数据中的点相比一个时间点下采集的三维点云数据中的点的个数会增多，因此，可以确定针对目标区域的拼接点云数据，以便后续基于拼接点云数据对不再准确的外部参数信息进行调整。Exemplarily, the target area refers to the same location area. After the world coordinate system is set in advance, the coordinate values of each location point in the target location area in the world coordinate system will not occur with the driving of the driving device. Therefore, the points used to represent the same position in the target area in the 3D point cloud data collected at different time points will overlap after splicing the 3D point cloud data when the external parameter information is accurate. In the case of inaccurate parameter information, the number of points in the spliced point cloud data obtained after splicing will increase compared to the number of points in the 3D point cloud data collected at one time point. Therefore, the splicing point for the target area can be determined. Cloud data, so that the no longer accurate external parameter information can be adjusted later based on the spliced point cloud data.

当然，在行驶设备行驶过程中，随着行驶设备位置的变化，在经过一段时间后，设置于行驶设备上的组合惯性导航设备和雷达传感器，获取到的目标区域即会发生变化，比如，在9:00～9:05之间获取到的是目标区域A，可以基于目标区域A对应的拼接点云数据来调整外部参数信息，在9:05～9:10之间获取到的是目标区域B，此时，可以基于目标区域B对应的拼接点云数据来调整外部参数信息，即在行驶设备的行驶过程中，可以按照不断获取到的目标区域对外部参数信息进行不断调整，本公开实施例仅针对其中一个目标区域对应拼接点云数据为例进行说明。Of course, during the driving process of the driving equipment, with the change of the position of the driving equipment, after a period of time, the combined inertial navigation equipment and radar sensor set on the driving equipment will change the acquired target area. For example, in the The target area A is obtained between 9:00 and 9:05. The external parameter information can be adjusted based on the splicing point cloud data corresponding to the target area A. The target area is obtained between 9:05 and 9:10. B, at this time, the external parameter information can be adjusted based on the splicing point cloud data corresponding to the target area B, that is, during the driving process of the driving device, the external parameter information can be continuously adjusted according to the continuously obtained target area. The example only takes the splicing point cloud data corresponding to one of the target regions as an example for description.

S104，基于拼接点云数据，对外部参数信息进行调整，得到目标外部参数信息。S104, based on the spliced point cloud data, adjust the external parameter information to obtain target external parameter information.

示例性地，同一目标区域的同一个位置点，在不同时间点下在世界坐标系中的位置信息应该相同。但是若初始的外部参数信息不准确，或者随着行驶设备的行驶，初始的外部参数信息不再准确时，会造成在不同时间点得到的同一目标区域中相同的位置点的位置信息不再相同，这样在对不同时间点对应的三维点云数据进行拼接时，会使得原本应该重合的位置点不再重合，即拼接点云数据中的点的个数相比单个时间点得到的三维点云数据中的点的个数会变多。基于此，可以基于拼接点云数据中点的个数对外部参数信息进行调整，从而对外部参数信息进行不断优化，最终得到准确度较高的目标外部参数信息。Exemplarily, for the same location point in the same target area, the location information in the world coordinate system at different time points should be the same. However, if the initial external parameter information is inaccurate, or the initial external parameter information is no longer accurate as the driving equipment is running, the position information of the same location point in the same target area obtained at different time points will no longer be the same. , in this way, when splicing the 3D point cloud data corresponding to different time points, the position points that should be coincident will no longer overlap, that is, the number of points in the spliced point cloud data is compared with the 3D point cloud obtained at a single time point. The number of points in the data will increase. Based on this, the external parameter information can be adjusted based on the number of points in the spliced point cloud data, so as to continuously optimize the external parameter information, and finally obtain the target external parameter information with high accuracy.

针对上述提到的S101～S104，在通过预先确定的外部参数信息确定出雷达传感器在不同时间点的位姿数据后，这样可以结合雷达传感器在不同时间点采集的三维点云数据，确定出不同时间点下表征目标区域的三维点云数据，在外部参数信息不再准确的情况下，不同时间点下表征目标区域同一位置点的坐标值不再相同，使得拼接点云数据中的点变多，基于此，可以基于拼接点云数据对外部参数信息进行调整，从而不断优化外部参数信息，直至得到准确度较高的目标外部参数信息。For S101 to S104 mentioned above, after determining the pose data of the radar sensor at different time points through the predetermined external parameter information, it is possible to combine the three-dimensional point cloud data collected by the radar sensor at different time points to determine different For the 3D point cloud data representing the target area at a time point, when the external parameter information is no longer accurate, the coordinate values of the same point representing the target area at different time points are no longer the same, making the spliced point cloud data more points. , based on this, the external parameter information can be adjusted based on the spliced point cloud data, so as to continuously optimize the external parameter information until the target external parameter information with higher accuracy is obtained.

下面将结合具体实施例，针对上述S101～S104进行详细阐述。The foregoing S101 to S104 will be described in detail below with reference to specific embodiments.

针对上述S102，在基于表征雷达传感器与组合惯性导航设备之间坐标系转换关系的外部参数信息、以及组合惯性导航设备的位姿数据，确定雷达传感器的位姿数据时，可以包括：For the above S102, when determining the pose data of the radar sensor based on the external parameter information representing the coordinate system conversion relationship between the radar sensor and the combined inertial navigation device, and the pose data of the combined inertial navigation device, it may include:

(1)基于外部参数信息、以及组合惯性导航设备在各个第一时间点的位姿数据，确定雷达传感器在各个第一时间点的位姿数据。(1) Determine the pose data of the radar sensor at each first time point based on the external parameter information and the pose data of the combined inertial navigation device at each first time point.

示例性地，可以通过外部参数信息以及组合惯性导航设备在各个第一时间点下在世界坐标系下的位姿数据，确定出雷达传感器在各个第一时间点下在世界坐标系中的位姿数据。Exemplarily, the pose of the radar sensor in the world coordinate system at each first time point can be determined by using the external parameter information and the pose data of the combined inertial navigation device in the world coordinate system at each first time point. data.

(2)按照各个第二时间点，对雷达传感器在各个第一时间点的位姿数据进行插值处理，并抽取插值处理后的位姿数据，得到雷达传感器在各个第二时间点的位姿数据。(2) Perform interpolation processing on the pose data of the radar sensor at each first time point according to each second time point, and extract the interpolated pose data to obtain the pose data of the radar sensor at each second time point .

因为第一时间点和第二时间点可能并不相同，这里在得到雷达传感器在各个第一时间点的位姿数据后，可以按照各个第二时间点，对雷达传感器在各个第一时间点的位姿数据进行差值处理，从而得到雷达传感器在各个第二时间点的位姿数据。Because the first time point and the second time point may not be the same, after obtaining the pose data of the radar sensor at each first time point, the radar sensor at each first time point can be analyzed according to each second time point. The pose data is subjected to difference processing, thereby obtaining pose data of the radar sensor at each second time point.

比如，各个第一时间点为从t时刻开始后的第1s、第3s、第5s、第7s…，而各个第二时间点为从t时刻开始后的第2s、第4s、第6s、第8s…，则上述过程是先基于外部参数信息、以及组合惯性导航设备在第1s、第3s、第5s、第7s…的位姿数据，确定出雷达传感器在第1s、第3s、第5s、第7s…的位姿数据，然后对雷达传感器在第1s、第 3s、第5s、第7s…的位姿数据进行插值处理，得到雷达传感器在第2s、第4s、第6s、第8s…时的位姿数据。For example, each first time point is the 1s, 3s, 5s, 7s after starting from time t, and each second time point is the 2s, 4s, 6s, 7s after starting from time t 8s..., the above process is based on the external parameter information and the pose data of the combined inertial navigation equipment at the 1s, 3s, 5s, 7s... The pose data of the 7th..., then the radar sensor's pose data at the 1s, 3s, 5s, 7s... pose data.

上述过程，可以通过预先确定的外部参数信息和组合惯性导航设备采集的该组合惯性导航设备在多个时间点下的位姿数据，得到雷达传感器在相应时间点下的位姿数据，从而便于进一步结合雷达传感器在相应时间点下采集的三维点云数据得到拼接点云数据。In the above process, the position and attitude data of the radar sensor at the corresponding time point can be obtained through the predetermined external parameter information and the position and attitude data of the combined inertial navigation device at multiple time points collected by the combined inertial navigation device, so as to facilitate further steps. Combined with the three-dimensional point cloud data collected by the radar sensor at the corresponding time point, the stitched point cloud data is obtained.

上述提到，获取到的雷达传感器在各个第二时间点采集的三维点云数据可以是针对获取到的雷达传感器在各个第二时间点传输的三维点云数据进行去畸变处理后得到的，示例性地，可以在获取到雷达传感器在各个第二时间点的位姿数据、以及在各个第二时间点传输的三维点云数据后，可以按照以下方式确定雷达传感器在各个第二时间点采集的三维点云数据：As mentioned above, the acquired 3D point cloud data collected by the radar sensor at each second time point may be obtained by performing de-distortion processing on the acquired 3D point cloud data transmitted by the radar sensor at each second time point. For example, Specifically, after acquiring the pose data of the radar sensor at each second time point and the three-dimensional point cloud data transmitted at each second time point, the data collected by the radar sensor at each second time point can be determined in the following manner. 3D point cloud data:

对雷达传感器在各个第二时间点传输的三维点云数据进行去畸变处理，得到雷达传感器在各个第二时间点采集的三维点云数据，其中，雷达传感器在各个第二时间点采集的三维点云数据包含目标区域对应的三维点云数据中的点在各个第二时间点下相对于雷达传感器的距离信息。Perform de-distortion processing on the 3D point cloud data transmitted by the radar sensor at each second time point to obtain the 3D point cloud data collected by the radar sensor at each second time point, wherein the 3D point cloud data collected by the radar sensor at each second time point The cloud data includes distance information of points in the three-dimensional point cloud data corresponding to the target area relative to the radar sensor at each second time point.

示例性地，去畸变处理即针对雷达传感器的运动畸变处理，比如，雷达传感器每隔1s采集一次三维点云数据，每隔2s传输一次三维点云数据，这样雷达传感器在每个第二时间点传输的三维点云数据中可以包含该第二时间点和其它时间点采集到的三维点云数据中的点相对于雷达传感器的距离信息，因为雷达传感器在不断运动，即在上个时间点采集的三维点云数据中的点相对于雷达传感器的距离信息，在当前时间点相对于雷达传感器的距离信息会发生变化，鉴于此，为了准确地得到目标区域中的多个位置点在各个第二时间点下相对于雷达传感器的距离信息，采用对雷达传感器在各个第二时间点传输的三维点云数据进行去畸变处理。Exemplarily, the de-distortion processing is the motion distortion processing for the radar sensor. For example, the radar sensor collects 3D point cloud data every 1s, and transmits the 3D point cloud data every 2s, so that the radar sensor at every second time point. The transmitted 3D point cloud data may include the distance information of the points in the 3D point cloud data collected at the second time point and other time points relative to the radar sensor, because the radar sensor is constantly moving, that is, collected at the previous time point. The distance information of the points in the 3D point cloud data relative to the radar sensor will change at the current time point relative to the radar sensor. In view of this, in order to accurately obtain the multiple position points in the target area in each second For the distance information relative to the radar sensor at the time point, the three-dimensional point cloud data transmitted by the radar sensor at each second time point is de-distorted.

具体地，可以按照以下方式对雷达传感器在任一第二时间点传输的目标区域的三维点云数据进行去畸变处理：Specifically, the three-dimensional point cloud data of the target area transmitted by the radar sensor at any second time point can be de-distorted in the following manner:

将雷达传感器在任一第二时间点传输的三维点云数据中，在各个时间点采集的三维点云数据中的点相对于雷达传感器的距离信息，转换到在任一第二时间点下相对于雷达传感器的距离信息。In the 3D point cloud data transmitted by the radar sensor at any second time point, the distance information of the points in the 3D point cloud data collected at each time point relative to the radar sensor is converted to the distance information relative to the radar sensor at any second time point. Distance information from the sensor.

比如雷达传感器每隔2s传输并采集一次三维点云数据，每隔1s采集一次三维点云数据，若从9:00:00开始，该任一第二时间点为第一个第二时间点时，即9:00:02时传输的三维点云数据中包括9:00:01和9:00:02采集的三维点云数据，此时，9:00:01时采集的三维点云数据中的点相对于雷达传感器的距离信息可以记录为L1，从9:00:01到9:00:02的过程中，因行驶设备和雷达传感器的转动，采集到的三维点云数据中的点A在9:00:02时相对于雷达传感器的距离信息相比该点A在9:00:01时相对于雷达传感器的距离信息可能发生了变化，此时可以将雷达传感器采集到的三维点云数据中的点在9:00:01时相对于雷达传感器的距离信息，转换到在9:00:02时相对于雷达传感器的距离信息。For example, the radar sensor transmits and collects 3D point cloud data every 2s, and collects 3D point cloud data every 1s. If it starts from 9:00:00, any second time point is the first second time point , that is, the 3D point cloud data transmitted at 9:00:02 includes the 3D point cloud data collected at 9:00:01 and 9:00:02. At this time, the 3D point cloud data collected at 9:00:01 The distance information of the point relative to the radar sensor can be recorded as L1. During the process from 9:00:01 to 9:00:02, due to the rotation of the driving equipment and the radar sensor, point A in the collected 3D point cloud data Compared with the distance information relative to the radar sensor at 9:00:02, the distance information of the point A relative to the radar sensor at 9:00:01 may have changed. At this time, the three-dimensional point cloud collected by the radar sensor can be Points in the data have distance information relative to the radar sensor at 9:00:01, converted to distance information relative to the radar sensor at 9:00:02.

具体转换时，雷达传感器采集到的三维点云数据中的点在9:00:01时相对于雷达传感器的距离信息，与在9:00:02时相对于雷达传感器的距离信息之间的距离转换关系可以通过雷达传感器在9:00:01时的位姿数据与9:00:02时的位姿数据确定，然后基于该距离转换关系，将雷达传感器采集到的三维点云数据中的点在9:00:01时相对于雷达传感器的距离信息，转换到在9:00:02时相对于雷达传感器的距离信息。During specific conversion, the distance between the points in the 3D point cloud data collected by the radar sensor relative to the radar sensor at 9:00:01 and the distance information relative to the radar sensor at 9:00:02 The conversion relationship can be determined by the pose data of the radar sensor at 9:00:01 and the pose data at 9:00:02, and then based on the distance conversion relationship, the points in the three-dimensional point cloud data collected by the radar sensor are The distance information relative to the radar sensor at 9:00:01 is converted to the distance information relative to the radar sensor at 9:00:02.

通过以上方式，可以采集到同一目标区域更加全面的三维点云数据，因为雷达传感器在采集三维点云数据过程中可能不断调整位姿，这样在对目标区域进行三维点云数据采集时，不同采集时间点采集到的目标区域中的位置点在该目标区域中所属的局部区域可能不同，比如当目标区域为某交通路口的车道线时，若9:00:01时采集到的是车道线的左半区域对应的三维点云数据，9:00:02采集到的是车道线的右半区域对应的三维点云数据，通过将9:00:01时采集到的三维点云数据中的点相对于雷达传感器的距离信息，转换到在9:00:02时相对于雷达传感器的距离信息，可以得到9:00:02时，该交通路口的全部车道线对应的三维点云数据。Through the above methods, more comprehensive 3D point cloud data of the same target area can be collected, because the radar sensor may continuously adjust the pose during the process of collecting 3D point cloud data, so that when collecting 3D point cloud data for the target area, different collections The location points in the target area collected at the time point may belong to different local areas in the target area. For example, when the target area is the lane line of a traffic intersection, if the lane line is collected at 9:00:01. The 3D point cloud data corresponding to the left half area is collected at 9:00:02 is the 3D point cloud data corresponding to the right half area of the lane line. By comparing the points in the 3D point cloud data collected at 9:00:01 The distance information relative to the radar sensor is converted to the distance information relative to the radar sensor at 9:00:02, and the three-dimensional point cloud data corresponding to all lane lines at the traffic intersection at 9:00:02 can be obtained.

通过以上过程对雷达传感器在各个第二时间点传输的三维点云数据进行去畸变处理，可以准确得到三维点云数据中的点在各个第二时间点相对于雷达传感器的距离信息，这样结合雷达传感器在各个第二时间点时对应的位姿数据，可以准确的得到三维点云数据中的点在世界坐标系下的位置信息。Through the above process, the 3D point cloud data transmitted by the radar sensor at each second time point is de-distorted, and the distance information of the points in the 3D point cloud data relative to the radar sensor at each second time point can be accurately obtained. The position and attitude data corresponding to the sensor at each second time point can accurately obtain the position information of the point in the three-dimensional point cloud data in the world coordinate system.

具体地，在基于雷达传感器在不同时间点对应的位姿数据、以及在相应时间点采集的三维点云数据，确定对目标区域的三维点云数据进行拼接后得到的拼接点云数据时，包括：Specifically, when determining the spliced point cloud data obtained by splicing the 3D point cloud data of the target area based on the pose data corresponding to the radar sensor at different time points and the 3D point cloud data collected at the corresponding time points, including :

基于雷达传感器在不同时间点的位姿数据、以及在相应时间点采集的三维点云数据中的点相对于雷达传感器的距离信息，确定拼接点云数据。Based on the pose data of the radar sensor at different time points and the distance information of the points in the 3D point cloud data collected at the corresponding time point relative to the radar sensor, the spliced point cloud data is determined.

示例性地，可以先通过雷达传感器在不同时间点的位姿数据、以及三维点云数据中的点在相应时间点相对于雷达传感器的距离信息，确定出三维点云数据中的点在不同时间点下在世界坐标系中的位置信息，然后进一步基于该位置信息，对不同时间点对应的三维点云数据中的点进行拼接，具体如下：Exemplarily, the pose data of the radar sensor at different time points and the distance information of the points in the three-dimensional point cloud data relative to the radar sensor at the corresponding time points can be used to determine the points in the three-dimensional point cloud data at different times. Click the position information in the world coordinate system, and then further based on the position information, splicing the points in the 3D point cloud data corresponding to different time points, as follows:

(1)基于雷达传感器在不同时间点的位姿数据、以及在相应时间点采集的三维点云数据中的点相对于雷达传感器的距离信息，确定不同时间点下目标区域对应的三维点云数据中的各个点的位置信息；(1) Based on the pose data of the radar sensor at different time points and the distance information of the points in the 3D point cloud data collected at the corresponding time points relative to the radar sensor, determine the 3D point cloud data corresponding to the target area at different time points The location information of each point in ;

(2)基于不同时间点下目标区域对应的三维点云数据中的各个点的位置信息，对不同时间点下目标区域的三维点云数据中的点进行拼接，得到拼接点云数据。(2) Based on the position information of each point in the 3D point cloud data corresponding to the target area at different time points, the points in the 3D point cloud data of the target area at different time points are spliced to obtain spliced point cloud data.

这里雷达传感器在不同时间点的位姿数据，可以是上述提到的雷达传感器在各个第二时间点下在世界坐标系下的位姿数据，根据这些位姿数据即可以得到在各个第二时间点下雷达传感器在世界坐标系下的坐标值、以及该雷达传感器在该世界坐标系下的朝向，此时，在得到目标区域的三维点云数据中的点在各个第二时间点相对于雷达传感器的距离信息时，即可以确定各个第二时间点下，目标位置区域的三维点云数据中的各个点在世界坐标系中的坐标值。Here, the pose data of the radar sensor at different time points can be the pose data of the radar sensor mentioned above in the world coordinate system at each second time point. According to these pose data, it can be obtained at each second time The coordinate value of the radar sensor in the world coordinate system and the orientation of the radar sensor in the world coordinate system. At this time, the points in the three-dimensional point cloud data of the target area are obtained relative to the radar at each second time point. When the distance information of the sensor is obtained, the coordinate value of each point in the world coordinate system in the three-dimensional point cloud data of the target location area at each second time point can be determined.

进一步地，可以基于各个第二时间点下，目标区域的三维点云数据中的各个点在世界坐标系下的坐标值，对目标区域在各个第二时间点对应的三维点云数据中的点进行拼接，比如，针对5个第二时间点，可以得到5组三维点云数据，每组三维点云数据均包含各个点在世界坐标系下的坐标值，然后对着5组三维点云数据进行拼接，坐标值相同的两个点可以重合变为一个点。Further, based on the coordinate values of each point in the three-dimensional point cloud data of the target area in the world coordinate system at each second time point, the points in the three-dimensional point cloud data corresponding to the target area at each second time point can be analyzed. For splicing, for example, for 5 second time points, 5 sets of 3D point cloud data can be obtained, each set of 3D point cloud data contains the coordinate values of each point in the world coordinate system, and then 5 sets of 3D point cloud data can be obtained. For splicing, two points with the same coordinate value can be overlapped to become one point.

特别地，为了降低数据冗余，基于所述不同时间点下所述目标区域对应的三维点云数据中的各个点的位置信息，对不同时间点下所述目标区域对应的三维点云数据中的点进行拼接，得到所述拼接点云数据，包括：In particular, in order to reduce data redundancy, based on the position information of each point in the 3D point cloud data corresponding to the target area at the different time points, the 3D point cloud data corresponding to the target area at different time points The points are spliced to obtain the spliced point cloud data, including:

(1)基于所述不同时间点下所述目标区域对应的三维点云数据中的各个点的位置信息，对所述目标区域对应的三维点云数据中位置重复的点进行去重处理；(1) Based on the position information of each point in the three-dimensional point cloud data corresponding to the target area at the different time points, deduplication processing is performed on the duplicated points in the three-dimensional point cloud data corresponding to the target area;

(2)基于进行所述去重处理后、所述不同时间点下所述目标区域对应的三维点云数据中的各个点的位置信息，对不同时间点下所述目标区域对应的三维点云数据中的点进行拼接，得到所述拼接点云数据。(2) Based on the position information of each point in the 3D point cloud data corresponding to the target area at the different time points after the deduplication processing is performed, the 3D point cloud corresponding to the target area at different time points is analyzed. The points in the data are spliced to obtain the spliced point cloud data.

示例性地，位置重复的点是指在世界坐标系中坐标值相同的点，因为这些点在世界坐标系中的坐标值相同，拼接后在对应的图像中仅会出现一个点，因此，可以对坐标值相同的点进行去重处理，仅保留位置重复的点中的一个点对应的位置信息即可。Exemplarily, points with repeated positions refer to points with the same coordinate value in the world coordinate system. Because these points have the same coordinate value in the world coordinate system, only one point will appear in the corresponding image after splicing. Therefore, you can The points with the same coordinate value are deduplicated, and only the position information corresponding to one of the points whose positions are repeated can be retained.

在得到拼接点云数据后，下面介绍如何基于拼接点云数据，对外部参数信息进行调整，得到目标外部参数信息，如图2所示，可以包括以下S201～S203：After obtaining the spliced point cloud data, the following describes how to adjust the external parameter information based on the spliced point cloud data to obtain the target external parameter information, as shown in Figure 2, which may include the following S201 to S203:

S201，根据拼接点云数据，对外部参数信息进行调整。S201: Adjust external parameter information according to the spliced point cloud data.

示例性地，拼接点云数据可以是初始的拼接点云数据，也可以是基于调整过至少一次的外部参数信息确定的拼接点云数据；外部参数信息可以为初始的外部参数信息，也可以是进行过至少一次调整后的外部参数信息。Exemplarily, the splicing point cloud data may be initial splicing point cloud data, or may be splicing point cloud data determined based on external parameter information that has been adjusted at least once; the external parameter information may be initial external parameter information, or may be External parameter information after at least one adjustment.

具体地，外部参数信息可以包含多个外部参数，具体地，在根据拼接点云数据，对外部参数信息进行调整时，可以包括：Specifically, the external parameter information may include multiple external parameters. Specifically, when adjusting the external parameter information according to the spliced point cloud data, the external parameter information may include:

对外部参数信息中的至少一个外部参数进行调整，得到调整后的外部参数信息。Adjust at least one external parameter in the external parameter information to obtain the adjusted external parameter information.

示例性地，外部参数信息包含的外部参数可以包含上述提到的X，Y，Z，Roll，Pitch，Yaw，对外部参数信息的调整，可以认为对至少一个外部参数的调整。Exemplarily, the external parameters included in the external parameter information may include the above-mentioned X, Y, Z, Roll, Pitch, and Yaw. The adjustment of the external parameter information may be considered as the adjustment of at least one external parameter.

本公开实施例中，在外部参数信息包含多个外部参数的情况下，可以基于至少一个外部参数对外部参数信息进行细微的调整，从而得到准确度较高的目标外部参数信息。In the embodiment of the present disclosure, when the external parameter information includes multiple external parameters, the external parameter information may be finely adjusted based on at least one external parameter, thereby obtaining target external parameter information with high accuracy.

S202，确定调整后的外部参数信息与调整前的外部参数信息不同，基于调整后的外部参数信息重新确定拼接点云数据后，返回根据拼接点云数据，对外部参数信息进行调整的过程，直至调整后的外部参数信息与调整前的外部参数信息相同。S202, it is determined that the external parameter information after adjustment is different from the external parameter information before adjustment, and after re-determining the splicing point cloud data based on the adjusted external parameter information, return to the process of adjusting the external parameter information according to the splicing point cloud data, until The external parameter information after adjustment is the same as the external parameter information before adjustment.

示例性地，在对外部参数信息进行调整优化过程中，可能无法进行一次调整得到最准确的目标外部参数信息，此时可以对外部参数信息进行多次优化，在优化过程中，可以基于调整后的外部参数信息与调整前的外部参数信息是否相同，来确定是否达到优化截止条件，在达到优化截止条件后，可以认为拼接点云数据中的点的个数已经达到较少的情况，此时再停止对外部参数信息的优化。Exemplarily, in the process of adjusting and optimizing the external parameter information, it may not be possible to perform one adjustment to obtain the most accurate target external parameter information. At this time, the external parameter information can be optimized multiple times. Whether the external parameter information is the same as the external parameter information before adjustment is used to determine whether the optimization cut-off condition is reached. After reaching the optimization cut-off condition, it can be considered that the number of points in the spliced point cloud data has reached a small number. At this time Then stop the optimization of the external parameter information.

S203，将调整后的外部参数信息作为目标外部参数信息。S203, taking the adjusted external parameter information as target external parameter information.

在确定拼接点云数据的点的个数在优化过程中已经达到较少的数量，无需再对外部参数信息进行继续优化的情况下，可以将调整后的外部参数信息作为目标外部参数信息。When it is determined that the number of points for splicing point cloud data has reached a small number in the optimization process, and there is no need to continue optimizing the external parameter information, the adjusted external parameter information can be used as the target external parameter information.

本公开实施例中，考虑到在外部参数信息不再准确的情况下，某个时间点下目标区域对应的拼接点云数据中的点的个数会相比同一个时间点下的三维点云数据中的点的个数会变多，因此，可以基于拼接点云数据对外部参数信息进行调整，直至调整后的外部参数信息与调整前的外部参数信息大致相同时，可以认为得到了准确度较高的目标外部参数信息。In the embodiment of the present disclosure, considering that when the external parameter information is no longer accurate, the number of points in the spliced point cloud data corresponding to the target area at a certain time point will be compared with the 3D point cloud at the same time point The number of points in the data will increase. Therefore, the external parameter information can be adjusted based on the spliced point cloud data. When the adjusted external parameter information is approximately the same as the pre-adjusted external parameter information, it can be considered that the accuracy has been obtained. Higher target extrinsic parameter information.

具体地，在对外部参数信息中的至少一个外部参数进行调整，得到调整后的外部参数信息时，如图3所示，可以包括以下步骤S301～S308：Specifically, when at least one external parameter in the external parameter information is adjusted to obtain the adjusted external parameter information, as shown in FIG. 3 , the following steps S301 to S308 may be included:

S301，在对外部参数信息进行当前轮调整过程中，从多个外部参数中选择当前轮还未调整的外部参数；其中，对外部参数信息进行一轮调整包括对外部参数信息中的各个外部参数均进行调整。S301, in the process of adjusting the external parameter information in the current round, select an external parameter that has not been adjusted in the current round from a plurality of external parameters; wherein, performing one round of adjustment on the external parameter information includes adjusting each external parameter in the external parameter information are adjusted.

示例性地，这里对外部参数信息进行一轮调整，可以是针对外部参数信息中所有的外部参数进行调整，比如针对外部参数信息中X，Y，Z，Roll，Pitch，Yaw 6个外部参数均进行调整完毕即执行完一轮调整，特别地，这里对外部参数进行调整后，外部参数的参数值在调整前后可能并未发生变化，详见下文。Exemplarily, one round of adjustment to the external parameter information here may be to adjust all the external parameters in the external parameter information, for example, for the six external parameters of X, Y, Z, Roll, Pitch, and Yaw in the external parameter information. After the adjustment is completed, a round of adjustment is performed. In particular, after the external parameters are adjusted here, the parameter values of the external parameters may not change before and after the adjustment, as detailed below.

具体地，在对外部参数信息进行当前轮调整中，可以按照X，Y，Z，Roll，Pitch，Yaw的顺序，依次选择待进行调整的外部参数。Specifically, in the current round of adjustment of the external parameter information, the external parameters to be adjusted may be selected in sequence according to the order of X, Y, Z, Roll, Pitch, and Yaw.

S302，对选择的外部参数的参数值进行当前次调整，得到当前调整后的外部参数信息。S302: Perform the current current adjustment on the parameter value of the selected external parameter to obtain currently adjusted external parameter information.

示例性地，当前次调整可以是针对选择的外部参数进行初次调整，也可以是进行多次调整，以当前轮调整过程中，针对选择的目标外部参数Roll进行i次调整为例，可以按照以下调整方式进行调整，确定对该目标外部参数Roll进行调整时的调整步长和调整方向，然后按照以下公式进行调整：Exemplarily, the current adjustment may be an initial adjustment for the selected external parameter, or may be adjusted multiple times. Taking i times of adjustment for the selected target external parameter Roll during the current round of adjustment as an example, it can be performed as follows: Adjust the adjustment method, determine the adjustment step size and adjustment direction when adjusting the target external parameter Roll, and then adjust according to the following formula:

α _i＝α _i-1+dλ。 α _i =α _i-1 +dλ.

其中，λ表示选择的任一外部参数对应的调整步长；d表示调整方向，具体包括+1和-1两种情况，α _i-1表示选择的目标外部参数Roll在i次调整前的参数值；α _i表示选择的目标外部参数Roll在i次调整后的参数值。 Among them, λ represents the adjustment step size corresponding to any selected external parameter; d represents the adjustment direction, including two cases of +1 and -1, and α _i-1 represents the selected target external parameter Roll before the i adjustment. value; α _i represents the parameter value of the selected target external parameter Roll after i times of adjustment.

具体在调整时，可以先按照正方向进行调整，在按照正方向进行调整后得到的当前调整后的外部参数信息无法使得拼接点云数据中点的个数减少的情况下，可以继续按照反方向进行调整；或者先按照反方向进行调整，在按照反方向进行调整后得到的当前调整后的外部参数信息无法使得拼接点云数据中点的个数减少的情况下，可以继续按照正方向进行调整，先按照正方向进行调整还是先按照反方向进行调整在此不进行限定。Specifically, when adjusting, you can first adjust in the positive direction. If the current adjusted external parameter information obtained after adjusting in the positive direction cannot reduce the number of points in the spliced point cloud data, you can continue to follow the reverse direction. Adjust; or first adjust in the opposite direction, and continue to adjust in the positive direction if the current adjusted external parameter information obtained after adjusting in the opposite direction cannot reduce the number of points in the spliced point cloud data , whether to adjust in the positive direction first or in the opposite direction is not limited here.

S303，确定基于当前调整后的外部参数信息得到的拼接点云数据中的点的个数与基于当前次调整前的外部参数信息得到的拼接点云数据中的点的个数相比是否变少；其中，当前次调整前的外部参数信息包括选择的外部参数在进行当前次调整之前的参数值。S303, determining whether the number of points in the spliced point cloud data obtained based on the currently adjusted external parameter information is less than the number of points in the spliced point cloud data obtained based on the external parameter information before the current adjustment ; wherein, the external parameter information before the current adjustment includes the parameter value of the selected external parameter before the current adjustment.

示例性地，在对选择的目标外部参数基于上述公式调整后，可以得到当前调整后的外部参数信息，此时需要基于当前调整后的外部参数信息按照上述S101～S103的方式重新确定出目标区域对应的拼接点云数据，并判断基于当前调整后的外部参数信息得到的拼接点云数据中的点的个数与基于当前次调整前的外部参数信息得到的拼接点云数据中的点的个数相比是否变少，以上述公式为例，即判断通过含α _i的外部参数信息得到的拼接点云数据中点的个数相比基于包含α _i-1的外部参数信息得到的拼接点云数据中点的个数是否变少。 Exemplarily, after the selected target external parameter is adjusted based on the above formula, the currently adjusted external parameter information can be obtained. At this time, it is necessary to re-determine the target area based on the currently adjusted external parameter information in the manner of S101 to S103 above. Corresponding splicing point cloud data, and determine the number of points in the splicing point cloud data obtained based on the currently adjusted external parameter information and the number of points in the splicing point cloud data obtained based on the external parameter information before the current adjustment. Take the above formula as an example, that is to judge whether the number of points in the splicing point cloud data obtained by the external parameter information containing α _i is compared with the splicing point obtained based on the external parameter information containing α _i-1 Whether the number of points in the cloud data decreases.

S304，若变少，对选择的外部参数的参数值进行更新后，返回对选择的外部参数值进行当前次调整的过程，即返回S302。S304, if it becomes less, after updating the parameter value of the selected external parameter, return to the process of performing the current adjustment on the selected external parameter value, that is, return to S302.

示例性地，若当前调整后的外部参数信息对应的拼接点云数据中的点的个数相比未调整之前对应的拼接点云数据中点的个数变少，则对选择的目标外部参数Roll的参数值进行更新后，比如将α _i-1更新为α _i后，并返回步骤S302，然后执行步骤S303后，继续确定基于当前调整后的外部参数信息得到的拼接点云数据中的点的个数是否变少，在对选择的目标外部参数的参数值进行当前次调整过程中，包括逐次增加调整步长，即d为+1，以及逐次减少调整步长，即d为-1的过程，直至得到对应拼接点云数据中点的个数最少的外部参数信息。 Exemplarily, if the number of points in the splicing point cloud data corresponding to the currently adjusted external parameter information is less than the number of points in the corresponding splicing point cloud data before the adjustment, then the selected target external parameter After the parameter value of Roll is updated, for example, after updating α _i-1 to α _i , return to step S302, and then perform step S303, continue to determine the points in the splicing point cloud data obtained based on the currently adjusted external parameter information Whether the number of s is reduced, in the current adjustment process of the parameter values of the selected target external parameters, including increasing the adjustment step size successively, that is, d is +1, and decreasing the adjustment step size successively, that is, d is -1. Process until the external parameter information with the least number of points in the corresponding spliced point cloud data is obtained.

S305，若未变少，保持选择的外部参数在当前次调整前的参数值，并回从多个外部参数中选择当前轮还未调整的外部参数的过程，即返回S201。S305, if it does not decrease, keep the parameter value of the selected external parameter before the current adjustment, and return to the process of selecting the external parameter that has not been adjusted in the current round from a plurality of external parameters, that is, return to S201.

示例性地，未变少是指无论逐次增加调整步长还是逐次减少调整步长的过程中，基于当前调整后的外部参数信息得到的拼接点云数据中的点的个数与基于当前次调整前的外部参数信息得到的拼接点云数据中的点的个数相比均未变少，即说明无需再对本次选择的目标外部参数进行调整或者已经对本次选择的目标外部参数调整完毕，此时保持选择的外部参数在当前次调整之前的参数值。比如无论正向调整还是反向调整后，本次选择的目标外部参数Roll的参数值由α _i-1得到α _i后，拼接点云数据中点的个数均不再变小，则外部参数信息中外部参数Roll的参数值仍为α _i-1，返回步骤S301，继续选择其它的外部参数进行调整。直至所有外部参数均执行过S301-S305的步骤，执行S306。 Exemplarily, the number of points in the spliced point cloud data obtained based on the external parameter information after the current adjustment is the same as the number of points in the spliced point cloud data obtained based on the current adjusted external parameter information in the process of increasing the adjustment step size or decreasing the adjustment step size successively. The number of points in the spliced point cloud data obtained from the previous extrinsic parameter information has not decreased compared with that of the previous extrinsic parameter information, which means that there is no need to adjust the external parameters of the target selected this time or the external parameters of the target selected this time have been adjusted. , then keep the parameter value of the selected external parameter before the current adjustment. For example, after the forward adjustment or the reverse adjustment, after the parameter value of the target external parameter Roll selected this time is obtained from α _i _-1 , the number of points in the spliced point cloud data is no longer small, then the external parameter The parameter value of the external parameter Roll in the information is still α _i-1 , return to step S301, and continue to select other external parameters for adjustment. Step S306 is executed until all the external parameters have performed the steps of S301-S305.

S306，在对外部参数信息完成当前轮调整后，判断当前轮最后选择的外部参数的参数值在当前轮调整前后是否发生变化。S306, after completing the current round adjustment of the external parameter information, determine whether the parameter value of the external parameter finally selected in the current round has changed before and after the current round of adjustment.

示例性地，这里判断当前轮最后选择的外部参数值是否发生变化，即判断是否在当前轮调整过程中，对最后选择的外部参数的参数值是否进行了调整，若对该最后选择的外部参数的参数值进行了调整，说明在调整最后选择的外部参数的参数值后，拼接点云数据中点的个数会继续发生减少，否则，说明对该最后选择的外部参数的参数值进行了调整后，拼接点云数据不会再减少。Exemplarily, here it is judged whether the value of the external parameter selected last in the current round has changed, that is, whether the parameter value of the external parameter selected last has been adjusted during the adjustment process of the current round, if the external parameter selected last has been adjusted. The parameter value of , which means that after adjusting the parameter value of the last selected external parameter, the number of points in the spliced point cloud data will continue to decrease. Otherwise, it means that the parameter value of the last selected external parameter has been adjusted. After that, the stitched point cloud data will not be reduced any more.

S307，若发生变化，进行下一轮外部参数信息调整。S307, if there is a change, perform the next round of external parameter information adjustment.

S308，若未发生变化，确定对外部参数信息调整结果达到调整后的外部参数信息与调整前的外部参数信息相同的条件，将调整后的外部参数信息作为目标外部参数信息。S308 , if there is no change, it is determined that the adjustment result of the external parameter information reaches the same condition that the adjusted external parameter information is the same as the external parameter information before adjustment, and the adjusted external parameter information is used as the target external parameter information.

示例性地，按照上述S301～S308的过程，在针对外部参数信息进行当前轮调整过程中，若按照每个外部参数依次调整的方式，针对第6个外部参数的参数值进行调整时，发现得到的当前调整后的外部参数信息对应的拼接点云数据中的点的个数始终未变少，则说明当前轮最后选择的外部参数的参数值在对外部参数信息进行调整后未发生变化，即达到调整后的外部参数信息与调整前的外部参数信息相同的条件，可以认为调整后的外部参数信息对应的准确度可以达到要求，此时可以将调整后的外部参数信息可以作为目标外部参数信息。Exemplarily, according to the above-mentioned processes of S301 to S308, in the process of adjusting the current round of external parameter information, if the parameter value of the sixth external parameter is adjusted according to the method of adjusting each external parameter in turn, it is found that: The number of points in the splicing point cloud data corresponding to the currently adjusted external parameter information has not decreased, which means that the parameter value of the external parameter selected last in the current round has not changed after the external parameter information is adjusted, that is, It can be considered that the corresponding accuracy of the adjusted external parameter information can meet the requirements, and the adjusted external parameter information can be used as the target external parameter information. .

否则，若对第6个外部参数的参数值进行调整后，得到的当前调整后的外部参数信息对应的拼接点云数据中的点的个数仍然在减少，说明此时拼接点云数据中的点的个数还可以再继续减少，为了提高外部参数信息的准确性，还需要对外部参数信息进行下一轮调整，即返回步骤S301，按照该方式，重新调整外部参数信息，直至得到目标外部参数信息。Otherwise, if the parameter value of the sixth external parameter is adjusted, the number of points in the spliced point cloud data corresponding to the currently adjusted external parameter information is still decreasing, which means that the number of points in the spliced point cloud data is still decreasing. The number of points can be further reduced. In order to improve the accuracy of the external parameter information, the external parameter information needs to be adjusted in the next round, that is, return to step S301. In this way, the external parameter information is re-adjusted until the target external parameter information is obtained. Parameter information.

在一种实施方式中，在得到目标外部参数信息后，本公开实施例提供的标定方法还包括：In one embodiment, after obtaining the target external parameter information, the calibration method provided by the embodiment of the present disclosure further includes:

(1)获取行驶设备行驶过程中，组合惯性导航设备采集的位姿数据、以及雷达传感器采集的三维点云数据。(1) Obtain the pose data collected by the combined inertial navigation equipment and the three-dimensional point cloud data collected by the radar sensor during the driving process of the driving equipment.

具体地，在行驶设备行驶过程中，这里获取的组合惯性导航设备的位姿数据是指组合惯性导航设备在世界坐标系下的当前位姿数据；雷达传感器采集的三维点云数据包括三维点云数据中多个点在雷达坐标系下分别相对于雷达传感器的距离信息，具体详见上文，在此不再赘述。Specifically, during the driving process of the traveling device, the pose data of the integrated inertial navigation device obtained here refers to the current pose data of the integrated inertial navigation device in the world coordinate system; the three-dimensional point cloud data collected by the radar sensor includes the three-dimensional point cloud. The distance information of multiple points in the data relative to the radar sensor in the radar coordinate system is described in detail above, and will not be repeated here.

(2)基于目标外部参数信息和组合惯性导航设备的位姿数据，确定雷达传感器的位姿数据。(2) Determine the pose data of the radar sensor based on the target external parameter information and the pose data of the combined inertial navigation device.

这里确定雷达传感器的位姿数据与上文类似，在此不再赘述。Determining the pose data of the radar sensor here is similar to the above, and will not be repeated here.

(3)基于雷达传感器在不同时间点的位姿数据、以及在相应时间点采集的三维点云数据，确定三维点云数据中各个点的位置信息。(3) Determine the position information of each point in the three-dimensional point cloud data based on the pose data of the radar sensor at different time points and the three-dimensional point cloud data collected at the corresponding time point.

在得到雷达传感器在世界坐标系下的位姿数据后，即可以结合三维点云数据中各个点在雷达坐标系下相对于雷达传感器的距离信息，得到三维点云数据中各个点在世界坐标系中的位置信息。After obtaining the pose data of the radar sensor in the world coordinate system, the distance information of each point in the 3D point cloud data relative to the radar sensor in the radar coordinate system can be combined to obtain each point in the 3D point cloud data in the world coordinate system. location information in .

(4)基于三维点云数据中各个点的位置信息，构建行驶设备行驶的区域的地图。(4) Based on the position information of each point in the three-dimensional point cloud data, a map of the area where the traveling device travels is constructed.

行驶设备在设定区域行驶时，可以不断获取该设定区域中构成各个物体的三维点云数据，从而按照上述方式确定该设定区域中各个物体的形貌以及在该设定区域内的位置信息，即按照各个物体在设定区域内的位置信息，构建该设定区域对应的地图。When driving in a set area, the driving equipment can continuously obtain the three-dimensional point cloud data of each object in the set area, so as to determine the shape of each object in the set area and the position in the set area according to the above method. information, that is, according to the position information of each object in the set area, a map corresponding to the set area is constructed.

在一种实施方式中，在行驶设备行驶的区域的地图之后，本公开实施例提供的标定方法还包括：In one embodiment, after the map of the area where the traveling device travels, the calibration method provided by the embodiment of the present disclosure further includes:

(1)获取组合惯性导航设备采集的当前位姿数据、以及雷达传感器采集的三维点云数据中的多个点相对于雷达传感器的当前距离信息以及三维点云数据中的多个点的位置信息。(1) Obtain the current pose data collected by the integrated inertial navigation device, the current distance information of multiple points in the three-dimensional point cloud data collected by the radar sensor relative to the radar sensor, and the position information of multiple points in the three-dimensional point cloud data. .

这里需要得到行驶设备行驶的区域中目标物对应的三维点数据中的至少三个不同点相对于雷达传感器的当前距离信息、以及这些点在预先建立的世界坐标系中的位置信息，具体地，这些点的位置信息可以是在构建地图后得到并保存的。Here, it is necessary to obtain the current distance information of at least three different points in the three-dimensional point data corresponding to the target in the area where the traveling device is traveling relative to the radar sensor, and the position information of these points in the pre-established world coordinate system. Specifically, The location information of these points can be obtained and saved after the map is constructed.

具体地，这里组合惯性导航设备采集的当前位姿数据是指在世界坐标系下的位姿数据；雷达传感器采集的三维点云数据中的多个点相对于雷达传感器的当前距离信息是指在雷达坐标系下相对于雷达传感器的当前距离信息。Specifically, the current pose data collected by the combined inertial navigation device here refers to the pose data in the world coordinate system; the current distance information of multiple points in the three-dimensional point cloud data collected by the radar sensor relative to the radar sensor refers to the position and attitude data in the world coordinate system. Current distance information relative to the radar sensor in the radar coordinate system.

(2)基于三维点云数据中的多个点相对于雷达传感器的当前距离信息以及三维点云数据中的多个点的位置信息，确定雷达传感器的当前位姿数据。(2) Determine the current pose data of the radar sensor based on the current distance information of the multiple points in the three-dimensional point cloud data relative to the radar sensor and the position information of the multiple points in the three-dimensional point cloud data.

基于各个点在世界坐标系中的位置信息、以及各个点在雷达坐标系下相对于雷达传感器的当前距离信息，即可以估计雷达传感器在世界坐标系下的当前位姿数据。Based on the position information of each point in the world coordinate system and the current distance information of each point relative to the radar sensor in the radar coordinate system, the current pose data of the radar sensor in the world coordinate system can be estimated.

(3)基于组合惯性导航设备的当前位姿数据以及雷达传感器的当前位姿数据，确定行驶设备的当前位姿数据。(3) Determine the current pose data of the traveling device based on the current pose data of the combined inertial navigation device and the current pose data of the radar sensor.

这里即可以将组合惯性导航设备的当前位姿数据和雷达传感器的当前位姿数据进行结合，综合得到行驶设备的当前位姿数据，即完成了对行驶设备的精准定位。Here, the current pose data of the combined inertial navigation device and the current pose data of the radar sensor can be combined, and the current pose data of the traveling device can be comprehensively obtained, that is, the precise positioning of the traveling device is completed.

本领域技术人员可以理解，在具体实施方式的上述方法中，各步骤的撰写顺序并不意味着严格的执行顺序而对实施过程构成任何限定，各步骤的具体执行顺序应当以其功能和可能的内在逻辑确定。Those skilled in the art can understand that in the above method of the specific implementation, the writing order of each step does not mean a strict execution order but constitutes any limitation on the implementation process, and the specific execution order of each step should be based on its function and possible Internal logic is determined.

基于同一技术构思，本公开实施例中还提供了与标定方法对应的标定装置，由于本公开实施例中的装置解决问题的原理与本公开实施例上述标定方法相似，因此装置的实施可以参见方法的实施，重复之处不再赘述。Based on the same technical concept, the embodiment of the present disclosure also provides a calibration device corresponding to the calibration method. Since the principle of solving the problem of the device in the embodiment of the present disclosure is similar to the above-mentioned calibration method in the embodiment of the present disclosure, the implementation of the device can refer to the method of implementation, and the repetition will not be repeated.

参照图4所示，本公开实施例提供了一种标定装置，该标定装置400包括：获取模块401、确定模块402、拼接模块403和调整模块404。Referring to FIG. 4 , an embodiment of the present disclosure provides a calibration device. The calibration device 400 includes an acquisition module 401 , a determination module 402 , a splicing module 403 and an adjustment module 404 .

获取模块401，用于获取在行驶设备行驶过程中，行驶设备上的组合惯性导航设备的位姿数据、以及行驶设备上的雷达传感器采集的三维点云数据。The obtaining module 401 is configured to obtain the pose data of the integrated inertial navigation device on the traveling device and the three-dimensional point cloud data collected by the radar sensor on the traveling device during the traveling process of the traveling device.

确定模块402，用于基于表征雷达传感器与组合惯性导航设备之间坐标系转换关系的外部参数信息、以及组合惯性导航设备的位姿数据，确定雷达传感器的位姿数据。The determination module 402 is configured to determine the pose data of the radar sensor based on the external parameter information representing the coordinate system transformation relationship between the radar sensor and the integrated inertial navigation device, and the pose data of the integrated inertial navigation device.

拼接模块403，用于基于雷达传感器在不同时间点的位姿数据、以及在相应时间点采集的三维点云数据，确定对目标区域对应的三维点云数据进行拼接后得到的拼接点云数据。The splicing module 403 is configured to determine spliced point cloud data obtained by splicing the 3D point cloud data corresponding to the target area based on the pose data of the radar sensor at different time points and the 3D point cloud data collected at the corresponding time points.

调整模块404，用于基于拼接点云数据，对外部参数信息进行调整，得到目标外部参数信息。The adjustment module 404 is configured to adjust the external parameter information based on the spliced point cloud data to obtain target external parameter information.

在一种可能的实施方式中，拼接模块403用于：In a possible implementation, the splicing module 403 is used for:

基于雷达传感器在不同时间点的位姿数据、以及在相应时间点采集的三维点云数据中的点相对于雷达传感器的距离信息，确定不同时间点下，目标区域的三维点云数据中的各个点的位置信息；Based on the pose data of the radar sensor at different time points and the distance information of the points in the 3D point cloud data collected at the corresponding time points relative to the radar sensor, determine the 3D point cloud data of the target area at different time points. location information of the point;

基于不同时间点下，目标区域的三维点云数据中的各个点的位置信息，对不同时间点下目标区域的三维点云数据中的点进行拼接，得到拼接点云数据。Based on the position information of each point in the 3D point cloud data of the target area at different time points, the points in the 3D point cloud data of the target area at different time points are spliced to obtain spliced point cloud data.

在一种可能的实施方式中，拼接模块403在基于所述不同时间点下所述目标区域对应的三维点云数据中的各个点的位置信息，在对不同时间点下所述目标区域对应的三维点云数据中的点进行拼接，得到所述拼接点云数据时，用于：In a possible implementation manner, the splicing module 403, based on the position information of each point in the three-dimensional point cloud data corresponding to the target area at the different time points, The points in the three-dimensional point cloud data are spliced, and when the spliced point cloud data is obtained, it is used for:

基于所述不同时间点下所述目标区域对应的三维点云数据中的各个点的位置信息，对所述目标区域对应的三维点云数据中位置重复的点进行去重处理；Based on the position information of each point in the 3D point cloud data corresponding to the target area at the different time points, deduplication processing is performed on the duplicated points in the 3D point cloud data corresponding to the target area;

基于进行所述去重处理后、所述不同时间点下所述目标区域对应的三维点云数据中的各个点的位置信息，对不同时间点下所述目标区域对应的三维点云数据中的点进行拼接，得到所述拼接点云数据。Based on the position information of each point in the 3D point cloud data corresponding to the target area at the different time points after the deduplication process is performed, for the 3D point cloud data corresponding to the target area at different time points The points are spliced to obtain the spliced point cloud data.

在一种可能的实施方式中，调整模块404用于：In one possible implementation, the adjustment module 404 is used to:

根据拼接点云数据，对外部参数信息进行调整；Adjust the external parameter information according to the splicing point cloud data;

确定调整后的外部参数信息与调整前的外部参数信息不同，基于调整后的外部参数信息重新确定拼接点云数据后，返回根据拼接点云数据，对外部参数信息进行调整的过程，直至调整后的外部参数信息与调整前的外部参数信息相同；Determine that the external parameter information after adjustment is different from the external parameter information before adjustment. After re-determining the splicing point cloud data based on the adjusted external parameter information, return to the process of adjusting the external parameter information according to the splicing point cloud data until after the adjustment The external parameter information is the same as the external parameter information before adjustment;

将调整后的外部参数信息作为目标外部参数信息。The adjusted extrinsic parameter information is used as the target extrinsic parameter information.

在一种可能的实施方式中，外部参数信息包含多个外部参数，调整模块404在用于根据拼接点云数据，对外部参数信息进行调整时，包括：In a possible implementation manner, the external parameter information includes multiple external parameters, and when the adjustment module 404 is used to adjust the external parameter information according to the spliced point cloud data, it includes:

在一种可能的实施方式中，调整模块404在用于对外部参数信息中的至少一个外部参数进行调整，得到调整后的外部参数信息时，包括：In a possible implementation manner, when the adjustment module 404 is used to adjust at least one external parameter in the external parameter information to obtain the adjusted external parameter information, the method includes:

在对外部参数信息进行当前轮调整过程中，从多个外部参数中选择当前轮还未调整的外部参数；其中，对外部参数信息进行一轮调整包括对外部参数信息中的各个外部参数均进行调整；In the process of adjusting the external parameter information in the current round, the external parameters that have not been adjusted in the current round are selected from a plurality of external parameters; wherein, performing one round of adjustment on the external parameter information includes adjusting each external parameter in the external parameter information. Adjustment;

对选择的外部参数的参数值进行当前次调整，得到当前调整后的外部参数信息；Perform the current adjustment on the parameter value of the selected external parameter to obtain the current adjusted external parameter information;

确定基于当前调整后的外部参数信息得到的拼接点云数据中的点的个数与基于当前次调整前的外部参数信息得到的拼接点云数据中的点的个数相比是否变少；其中，当前次调整前的外部参数信息包括选择的外部参数在进行当前次调整之前的参数值；Determine whether the number of points in the spliced point cloud data obtained based on the currently adjusted external parameter information is less than the number of points in the spliced point cloud data obtained based on the external parameter information before the current adjustment; wherein , the external parameter information before the current adjustment includes the parameter value of the selected external parameter before the current adjustment;

若变少，对选择的外部参数的参数值进行更新后，返回对选择的外部参数进行当前次调整的过程，否则，保持选择的外部参数在当前次调整前的参数值，返回从多个外部参数中选择当前轮还未调整的外部参数的过程；If it becomes less, after updating the parameter value of the selected external parameter, return to the process of performing the current adjustment of the selected external parameter, otherwise, keep the parameter value of the selected external parameter before the current adjustment, and return from multiple external parameters The process of selecting external parameters that have not been adjusted in the current round in the parameters;

在对外部参数信息完成当前轮调整后，判断当前轮最后选择的外部参数的参数值在当前轮调整前后是否发生变化；若发生变化，进行下一轮外部参数信息调整，若未发生变化，确定对外部参数信息的调整结果达到调整后的外部参数信息与调整前的外部参数信息相同的条件，将调整后的外部参数信息作为目标外部参数信息。After completing the current round of adjustment of the external parameter information, determine whether the parameter value of the last selected external parameter in the current round has changed before and after the current round of adjustment; The adjustment result of the external parameter information reaches the same condition as the external parameter information after adjustment and the external parameter information before adjustment, and the external parameter information after adjustment is used as the target external parameter information.

在一种可能的实施方式中，标定装置还包括构图模块405，在得到目标外部参数信息后，构图模块用于：In a possible implementation manner, the calibration device further includes a composition module 405, and after obtaining the target external parameter information, the composition module is used for:

获取行驶设备行驶过程中，组合惯性导航设备的位姿数据、以及雷达传感器采集的三维点云数据；Obtain the pose data of the combined inertial navigation equipment and the 3D point cloud data collected by the radar sensor during the driving process of the driving equipment;

基于目标外部参数信息和组合惯性导航设备的位姿数据，确定雷达传感器的位姿数据；Determine the pose data of the radar sensor based on the target external parameter information and the pose data of the combined inertial navigation equipment;

基于雷达传感器在不同时间点的位姿数据、以及在相应时间点采集的三维点云数据，确定在三维点云数据中各个点的位置信息；Determine the position information of each point in the 3D point cloud data based on the pose data of the radar sensor at different time points and the 3D point cloud data collected at the corresponding time point;

基于三维点云数据中各个点的位置信息，构建行驶设备行驶的区域的地图。Based on the position information of each point in the 3D point cloud data, a map of the area where the traveling device travels is constructed.

对应于图1中的标定方法，本公开实施例还提供了一种电子设备500，如图5所示，为本公开实施例提供的电子设备500结构示意图，包括：处理器51、存储器52和总线53。Corresponding to the calibration method in FIG. 1 , an embodiment of the present disclosure further provides an electronic device 500 . As shown in FIG. 5 , a schematic structural diagram of the electronic device 500 provided by the embodiment of the present disclosure includes: a processor 51 , a memory 52 and bus 53.

存储器52用于存储执行指令，包括内存521和外部存储器522；这里的内存521也称内存储器，用于暂时存放处理器51中的运算数据、以及与硬盘等外部存储器522交换的数据，处理器51通过内存521与外部存储器522进行数据交换，当电子设备500运行时，处理器51与存储器52之间通过总线53通信，使得处理器51执行以下指令：获取在行驶设备行驶过程中，行驶设备上的组合惯性导航设备的位姿数据、以及行驶设备上的雷达传感器采集的三维点云数据；基于表征雷达传感器与组合惯性导航设备之间坐标系转换关系的外部参数信息、以及组合惯性导航设备的位姿数据，确定雷达传感器的位姿数据；基于雷达传感器在不同时间点的位姿数据、以及在相应时间点采集的三维点云数据，确定对目标区域的三维点云数据进行拼接后得到的拼接点云数据；基于拼接点云数据，对外部参数信息进行调整，得到目标外部参数信息。The memory 52 is used to store the execution instructions, including the memory 521 and the external memory 522; the memory 521 here is also called the internal memory, and is used to temporarily store the operation data in the processor 51 and the data exchanged with the external memory 522 such as the hard disk. 51 performs data exchange with the external memory 522 through the memory 521. When the electronic device 500 is running, the processor 51 communicates with the memory 52 through the bus 53, so that the processor 51 executes the following instructions: Get the driving device during the driving process. The pose data of the integrated inertial navigation equipment on the integrated inertial navigation equipment, and the three-dimensional point cloud data collected by the radar sensor on the driving equipment; based on the external parameter information representing the coordinate system transformation relationship between the radar sensor and the integrated inertial navigation equipment, and the integrated inertial navigation equipment The pose data of the radar sensor is determined based on the pose data of the radar sensor at different time points and the 3D point cloud data collected at the corresponding time point, and the 3D point cloud data of the target area is determined after splicing. based on the splicing point cloud data; based on the splicing point cloud data, the external parameter information is adjusted to obtain the target external parameter information.

本公开实施例还提供一种计算机可读存储介质，该计算机可读存储介质上存储有计算机程序，该计算机程序被处理器运行时执行上述方法实施例中所述的标定方法的步骤。其中，该存储介质可以是易失性或非易失的计算机可读取存储介质。Embodiments of the present disclosure further provide a computer-readable storage medium, where a computer program is stored on the computer-readable storage medium, and when the computer program is run by a processor, the steps of the calibration method described in the foregoing method embodiments are executed. Wherein, the storage medium may be a volatile or non-volatile computer-readable storage medium.

本公开实施例还提供一种计算机程序产品，该计算机程序产品承载有程序代码，所述程序代码包括的指令可用于执行上述方法实施例中所述的标定方法的步骤，具体可参见上述方法实施例，在此不再赘述。Embodiments of the present disclosure further provide a computer program product, where the computer program product carries program codes, and the instructions included in the program codes can be used to execute the steps of the calibration method described in the foregoing method embodiments. For details, please refer to the foregoing method implementation. For example, it will not be repeated here.

其中，上述计算机程序产品可以具体通过硬件、软件或其结合的方式实现。在一个可选实施例中，所述计算机程序产品具体体现为计算机存储介质，在另一个可选实施例中，计算机程序产品具体体现为软件产品，例如软件开发包(Software Development Kit，SDK)等等。Wherein, the above-mentioned computer program product can be specifically implemented by means of hardware, software or a combination thereof. In an optional embodiment, the computer program product is embodied as a computer storage medium, and in another optional embodiment, the computer program product is embodied as a software product, such as a software development kit (Software Development Kit, SDK), etc. Wait.

所属领域的技术人员可以清楚地了解到，为描述的方便和简洁，上述描述的***和装置的具体工作过程，可以参考前述方法实施例中的对应过程，在此不再赘述。在本公开所提供的几个实施例中，应该理解到，所揭露的***、装置和方法，可以通过其它的方式实现。以上所描述的装置实施例仅仅是示意性的，例如，所述单元的划分，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式，又例如，多个单元或组件可以结合或者可以集成到另一个***，或一些特征可以忽略，或不执行。另一点，所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些通信接口，装置或单元的间接耦合或通信连接，可以是电性，机械或其它的形式。Those skilled in the art can clearly understand that, for the convenience and brevity of description, for the specific working process of the system and device described above, reference may be made to the corresponding process in the foregoing method embodiments, which will not be repeated here. In the several embodiments provided by the present disclosure, it should be understood that the disclosed system, apparatus and method may be implemented in other manners. The apparatus embodiments described above are only illustrative. For example, the division of the units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components may be combined or Can be integrated into another system, or some features can be ignored, or not implemented. On the other hand, the shown or discussed mutual coupling or direct coupling or communication connection may be through some communication interfaces, indirect coupling or communication connection of devices or units, which may be in electrical, mechanical or other forms.

所述作为分离部件说明的单元可以是或者也可以不是物理上分开的，作为单元显示的部件可以是或者也可以不是物理单元，即可以位于一个地方，或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本实施例方案的目的。The units described as separate components may or may not be physically separated, and components displayed as units may or may not be physical units, that is, may be located in one place, or may be distributed to multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution in this embodiment.

另外，在本公开各个实施例中的各功能单元可以集成在一个处理单元中，也可以是各个单元单独物理存在，也可以两个或两个以上单元集成在一个单元中。In addition, each functional unit in each embodiment of the present disclosure may be integrated into one processing unit, or each unit may exist physically alone, or two or more units may be integrated into one unit.

所述功能如果以软件功能单元的形式实现并作为独立的产品销售或使用时，可以存储在一个处理器可执行的非易失的计算机可读取存储介质中。基于这样的理解，本公开的技术方案本质上或者说对现有技术做出贡献的部分或者该技术方案的部分可以以软件产品的形式体现出来，该计算机软件产品存储在一个存储介质中，包括若干指令用以使得一台计算机设备(可以是个人计算机，服务器，或者网络设备等)执行本公开各个实施例所述方法的全部或部分步骤。而前述的存储介质包括：U盘、移动硬盘、只读存储器(Read-Only Memory，ROM)、随机存取存储器(Random Access Memory，RAM)、磁碟或者光盘等各种可以存储程序代码的介质。The functions, if implemented in the form of software functional units and sold or used as stand-alone products, may be stored in a processor-executable non-volatile computer-readable storage medium. Based on this understanding, the technical solutions of the present disclosure can be embodied in the form of software products in essence, or the parts that make contributions to the prior art or the parts of the technical solutions. The computer software products are stored in a storage medium, including Several instructions are used to cause a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present disclosure. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, ROM), random access memory (Random Access Memory, RAM), magnetic disk or optical disk and other media that can store program codes .

最后应说明的是：以上所述实施例，仅为本公开具体实施方式，用以说明本公开的技术方案，而非对其限制，本公开的保护范围并不局限于此，尽管参照前述实施例对本公开进行了详细说明，本领域的普通技术人员应当理解：任何熟悉本技术领域的技术人员在本公开揭露的技术范围内，其依然可以对前述实施例所记载的技术方案进行修改或可轻易想到变化，或者对其中部分技术特征进行等同替换；而这些修改、变化或者替换，并不使相应技术方案的本质脱离本公开实施例技术方案的精神和范围，都应涵盖在本公开的保护范围之内。因此，本公开的保护范围应所述以权利要求的保护范围为准。Finally, it should be noted that the above-mentioned embodiments are only specific implementations of the present disclosure, and are used to illustrate the technical solutions of the present disclosure, rather than limit them. The present disclosure is described in detail in the examples, and those of ordinary skill in the art should understand that: any person skilled in the art can still modify or modify the technical solutions described in the foregoing embodiments within the technical scope disclosed in the present disclosure. Changes are easily thought of, or equivalent replacements are made to some of the technical features; and these modifications, changes or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present disclosure, and should be included in the protection of the present disclosure. within the range. Therefore, the protection scope of the present disclosure should be based on the protection scope of the claims.

Claims

一种标定方法，包括：A calibration method, including:

获取在行驶设备行驶过程中，所述行驶设备上的组合惯性导航设备的位姿数据、以及所述行驶设备上的雷达传感器采集的三维点云数据；Acquire the pose data of the integrated inertial navigation device on the traveling device and the three-dimensional point cloud data collected by the radar sensor on the traveling device during the traveling process of the traveling device;

基于表征所述雷达传感器与所述组合惯性导航设备之间坐标系转换关系的外部参数信息、以及所述组合惯性导航设备的位姿数据，确定所述雷达传感器的位姿数据；Determine the pose data of the radar sensor based on the external parameter information representing the coordinate system transformation relationship between the radar sensor and the integrated inertial navigation device, and the pose data of the integrated inertial navigation device;

基于所述雷达传感器在不同时间点的位姿数据、以及在相应时间点采集的三维点云数据，确定对目标区域的三维点云数据进行拼接后得到的拼接点云数据；Based on the pose data of the radar sensor at different time points and the 3D point cloud data collected at the corresponding time points, determine the spliced point cloud data obtained by splicing the 3D point cloud data of the target area;

基于所述拼接点云数据对所述外部参数信息进行调整，得到目标外部参数信息。The external parameter information is adjusted based on the spliced point cloud data to obtain target external parameter information.
根据权利要求1所述的标定方法，其特征在于，所述基于所述雷达传感器在不同时间点的位姿数据、以及在相应时间点采集的三维点云数据，确定对目标区域的三维点云数据进行拼接后得到的拼接点云数据，包括：The calibration method according to claim 1, wherein the three-dimensional point cloud of the target area is determined based on the pose data of the radar sensor at different time points and the three-dimensional point cloud data collected at the corresponding time point. The spliced point cloud data obtained after the data is spliced, including:

基于所述雷达传感器在不同时间点的位姿数据、以及在相应时间点采集的三维点云数据中的点相对于所述雷达传感器的距离信息，确定所述拼接点云数据。The spliced point cloud data is determined based on the pose data of the radar sensor at different time points and the distance information of points in the three-dimensional point cloud data collected at the corresponding time points relative to the radar sensor.
根据权利要求2所述的标定方法，其特征在于，所述基于所述雷达传感器在不同时间点的位姿数据、以及在相应时间点采集的三维点云数据中的点相对于所述雷达传感器的距离信息，确定所述拼接点云数据，包括：The calibration method according to claim 2, characterized in that, the points in the three-dimensional point cloud data based on the pose data of the radar sensor at different time points and the three-dimensional point cloud data collected at corresponding time points are relative to the radar sensor The distance information to determine the splicing point cloud data, including:

基于所述雷达传感器在不同时间点的位姿数据、以及在相应时间点采集的三维点云数据中的点相对于所述雷达传感器的距离信息，确定所述不同时间点下，所述目标区域的三维点云数据中的各个点的位置信息；Based on the pose data of the radar sensor at different time points and the distance information of the points in the three-dimensional point cloud data collected at the corresponding time points relative to the radar sensor, determine the target area at the different time points The position information of each point in the 3D point cloud data;

基于所述不同时间点下，所述目标区域的三维点云数据中的各个点的位置信息，对不同时间点下所述目标区域的三维点云数据中的点进行拼接，得到所述拼接点云数据。Based on the position information of each point in the 3D point cloud data of the target area at the different time points, the points in the 3D point cloud data of the target area at different time points are spliced to obtain the splicing point cloud data.
根据权利要求3所述的标定方法，其特征在于，基于所述不同时间点下所述目标区域对应的三维点云数据中的各个点的位置信息，对不同时间点下所述目标区域对应的三维点云数据中的点进行拼接，得到所述拼接点云数据，包括：The calibration method according to claim 3, wherein, based on the position information of each point in the three-dimensional point cloud data corresponding to the target area at the different time points, The points in the three-dimensional point cloud data are spliced to obtain the spliced point cloud data, including:

基于所述不同时间点下所述目标区域对应的三维点云数据中的各个点的位置信息，对所述目标区域对应的三维点云数据中位置重复的点进行去重处理；Based on the position information of each point in the 3D point cloud data corresponding to the target area at the different time points, deduplication processing is performed on the duplicated points in the 3D point cloud data corresponding to the target area;

基于进行所述去重处理后、所述不同时间点下所述目标区域对应的三维点云数据中的各个点的位置信息，对不同时间点下所述目标区域对应的三维点云数据中的点进行拼接，得到所述拼接点云数据。Based on the position information of each point in the 3D point cloud data corresponding to the target area at the different time points after the deduplication process is performed, for the 3D point cloud data corresponding to the target area at different time points The points are spliced to obtain the spliced point cloud data.
根据权利要求1至4任一所述的标定方法，其特征在于，所述基于所述拼接点云数据，对所述外部参数信息进行调整，得到目标外部参数信息，包括：The calibration method according to any one of claims 1 to 4, wherein the external parameter information is adjusted based on the spliced point cloud data to obtain target external parameter information, comprising:

根据所述拼接点云数据，对所述外部参数信息进行调整；Adjusting the external parameter information according to the splicing point cloud data;

确定调整后的外部参数信息与调整前的外部参数信息不同，基于调整后的外部参数信息重新确定拼接点云数据后，返回根据所述拼接点云数据，对所述外部参数信息进行调整的过程，直至调整后的外部参数信息与调整前的外部参数信息相同；It is determined that the external parameter information after adjustment is different from the external parameter information before adjustment. After re-determining the splicing point cloud data based on the adjusted external parameter information, the process of adjusting the external parameter information according to the splicing point cloud data is returned. , until the external parameter information after adjustment is the same as the external parameter information before adjustment;

将调整后的所述外部参数信息作为所述目标外部参数信息。The adjusted external parameter information is used as the target external parameter information.
根据权利要求5所述的标定方法，其特征在于，所述外部参数信息包含多个外部参数，所述根据所述拼接点云数据，对所述外部参数信息进行调整，包括：The calibration method according to claim 5, wherein the external parameter information includes a plurality of external parameters, and the adjustment of the external parameter information according to the spliced point cloud data includes:

对所述外部参数信息中的至少一个外部参数进行调整，得到调整后的外部参数信息。Adjust at least one external parameter in the external parameter information to obtain adjusted external parameter information.
根据权利要求6所述的标定方法，其特征在于，所述对所述外部参数信息中的至少一个外部参数进行调整，得到调整后的外部参数信息，包括：The calibration method according to claim 6, wherein the adjusting at least one external parameter in the external parameter information to obtain the adjusted external parameter information, comprising:

在对所述外部参数信息进行当前轮调整过程中，从所述多个外部参数中选择当前轮还未调整的外部参数；其中，对外部参数信息进行一轮调整包括对所述外部参数信息中的各个外部参数均进行调整；During the current round of adjustment of the external parameter information, the external parameters that have not been adjusted in the current round are selected from the plurality of external parameters; wherein, the one-round adjustment of the external parameter information includes adjusting the external parameters in the external parameter information. All external parameters are adjusted;

对选择的外部参数的参数值进行当前次调整，得到当前调整后的外部参数信息；Perform the current adjustment on the parameter value of the selected external parameter to obtain the current adjusted external parameter information;

确定基于当前调整后的外部参数信息得到的拼接点云数据中的点的个数与基于当前次调整前的外部参数信息得到的拼接点云数据中的点的个数相比是否变少；其中，当前次调整前的外部参数信息包括选择的外部参数在进行当前次调整之前的参数值；若变少，对选择的外部参数的参数值进行更新后，返回对选择的外部参数进行当前次调整的过程，否则，保持选择的外部参数在当前次调整前的参数值，并返回从所述多个外部参数中选择当前轮还未调整的外部参数的过程；Determine whether the number of points in the spliced point cloud data obtained based on the currently adjusted external parameter information is less than the number of points in the spliced point cloud data obtained based on the external parameter information before the current adjustment; wherein , the external parameter information before the current adjustment includes the parameter value of the selected external parameter before the current adjustment; if it becomes less, after updating the parameter value of the selected external parameter, return to the current adjustment of the selected external parameter Otherwise, keep the parameter value of the selected external parameter before the current adjustment, and return to the process of selecting the external parameter that has not been adjusted in the current round from the multiple external parameters;

在对所述外部参数信息完成当前轮调整后，判断当前轮最后选择的外部参数的参数值在当前轮调整前后是否发生变化；若发生变化，进行下一轮外部参数信息调整，若未发生变化，确定对外部参数信息的调整结果达到调整后的外部参数信息与调整前的外部参数信息相同的条件，将调整后的外部参数信息作为所述目标外部参数信息。After completing the current round of adjustment of the external parameter information, determine whether the parameter value of the last selected external parameter in the current round has changed before and after the current round of adjustment; if there is a change, perform the next round of external parameter information adjustment. , determining that the adjustment result of the external parameter information reaches the same condition as the external parameter information after adjustment and the external parameter information before adjustment, and using the external parameter information after adjustment as the target external parameter information.
根据权利要求1至7任一所述的标定方法，其特征在于，在得到所述目标外部参数信息后，所述标定方法还包括：The calibration method according to any one of claims 1 to 7, wherein after obtaining the target external parameter information, the calibration method further comprises:

获取所述行驶设备行驶过程中，所述组合惯性导航设备的位姿数据、以及所述雷达传感器采集的三维点云数据；acquiring the pose data of the integrated inertial navigation device and the three-dimensional point cloud data collected by the radar sensor during the driving of the traveling device;

基于所述目标外部参数信息和所述组合惯性导航设备的位姿数据，确定所述雷达传感器的位姿数据；Determine the pose data of the radar sensor based on the target external parameter information and the pose data of the combined inertial navigation device;

基于所述雷达传感器在不同时间点的位姿数据、以及在相应时间点采集的三维点云数据，确定所述三维点云数据中各个点的位置信息；Determine the position information of each point in the three-dimensional point cloud data based on the pose data of the radar sensor at different time points and the three-dimensional point cloud data collected at the corresponding time points;

基于所述三维点云数据中各个点的位置信息，构建所述行驶设备行驶的区域的地图。Based on the position information of each point in the three-dimensional point cloud data, a map of the area where the traveling device travels is constructed.
一种标定装置，包括：A calibration device, comprising:

获取模块，用于获取在行驶设备行驶过程中，所述行驶设备上的组合惯性导航设备的位姿数据、以及所述行驶设备上的雷达传感器采集的三维点云数据；an acquisition module, configured to acquire the pose data of the integrated inertial navigation device on the traveling device and the three-dimensional point cloud data collected by the radar sensor on the traveling device during the traveling process of the traveling device;

确定模块，用于基于表征所述雷达传感器与所述组合惯性导航设备之间坐标系转换关系的外部参数信息、以及所述组合惯性导航设备的位姿数据，确定所述雷达传感器的位姿数据；A determination module, configured to determine the pose data of the radar sensor based on the external parameter information representing the coordinate system transformation relationship between the radar sensor and the integrated inertial navigation device, and the pose data of the integrated inertial navigation device ;

拼接模块，用于基于所述雷达传感器在不同时间点的位姿数据、以及在相应时间点采集的三维点云数据，确定对目标区域的三维点云数据进行拼接后得到的拼接点云数据；A splicing module, configured to determine spliced point cloud data obtained by splicing the three-dimensional point cloud data of the target area based on the pose data of the radar sensor at different time points and the three-dimensional point cloud data collected at the corresponding time points;

调整模块，用于基于所述拼接点云数据，对所述外部参数信息进行调整，得到目标外部参数信息。An adjustment module, configured to adjust the external parameter information based on the spliced point cloud data to obtain target external parameter information.
根据权利要求9所述的标定装置，其特征在于，所述拼接模块用于：The calibration device according to claim 9, wherein the splicing module is used for:

基于所述雷达传感器在不同时间点的位姿数据、以及在相应时间点采集的三维点云数据中的点相对于所述雷达传感器的距离信息，确定所述拼接点云数据。The spliced point cloud data is determined based on the pose data of the radar sensor at different time points and the distance information of points in the three-dimensional point cloud data collected at the corresponding time points relative to the radar sensor.
根据权利要求10所述的标定装置，其特征在于，所述拼接模块用于：The calibration device according to claim 10, wherein the splicing module is used for:

基于所述雷达传感器在不同时间点的位姿数据、以及在相应时间点采集的三维点云数据中的点相对于所述雷达传感器的距离信息，确定所述不同时间点下，所述目标区域的三维点云数据中的各个点的位置信息；Based on the pose data of the radar sensor at different time points and the distance information of the points in the three-dimensional point cloud data collected at the corresponding time points relative to the radar sensor, determine the target area at the different time points The position information of each point in the 3D point cloud data;

基于所述不同时间点下，所述目标区域的三维点云数据中的各个点的位置信息，对不同时间点下所述目标区域的三维点云数据中的点进行拼接，得到所述拼接点云数据。Based on the position information of each point in the 3D point cloud data of the target area at the different time points, the points in the 3D point cloud data of the target area at different time points are spliced to obtain the splicing point cloud data.
根据权利要求11所述的标定装置，其特征在于，所述拼接模块在基于所述不同时间点下所述目标区域对应的三维点云数据中的各个点的位置信息，在对不同时间点下所述目标区域对应的三维点云数据中的点进行拼接，得到所述拼接点云数据时，用于：The calibration device according to claim 11, wherein the splicing module, based on the position information of each point in the three-dimensional point cloud data corresponding to the target area at the different time points, The points in the three-dimensional point cloud data corresponding to the target area are spliced, and when the spliced point cloud data is obtained, it is used for:

基于所述不同时间点下所述目标区域对应的三维点云数据中的各个点的位置信息，对所述目标区域对应的三维点云数据中位置重复的点进行去重处理；Based on the position information of each point in the 3D point cloud data corresponding to the target area at the different time points, deduplication processing is performed on the duplicated points in the 3D point cloud data corresponding to the target area;

基于进行所述去重处理后、所述不同时间点下所述目标区域对应的三维点云数据中的各个点的位置信息，对不同时间点下所述目标区域对应的三维点云数据中的点进行拼接，得到所述拼接点云数据。Based on the position information of each point in the 3D point cloud data corresponding to the target area at the different time points after the deduplication process is performed, for the 3D point cloud data corresponding to the target area at different time points The points are spliced to obtain the spliced point cloud data.
根据权利要求9至12任一所述的标定装置，其特征在于，所述调整模块用于：The calibration device according to any one of claims 9 to 12, wherein the adjustment module is used for:

根据所述拼接点云数据，对所述外部参数信息进行调整；Adjusting the external parameter information according to the splicing point cloud data;

确定调整后的外部参数信息与调整前的外部参数信息不同，基于调整后的外部参数信息重新确定拼接点云数据后，返回根据所述拼接点云数据，对所述外部参数信息进行调整的过程，直至调整后的外部参数信息与调整前的外部参数信息相同；It is determined that the external parameter information after adjustment is different from the external parameter information before adjustment. After re-determining the splicing point cloud data based on the adjusted external parameter information, the process of adjusting the external parameter information according to the splicing point cloud data is returned. , until the external parameter information after adjustment is the same as the external parameter information before adjustment;

将调整后的外部参数信息作为所述目标外部参数信息。The adjusted external parameter information is used as the target external parameter information.
根据权利要求13所述的标定装置，其特征在于，所述外部参数信息包含多个外部参数，所述调整模块在用于根据所述拼接点云数据，对所述外部参数信息进行调整时，包括：The calibration device according to claim 13, wherein the external parameter information includes a plurality of external parameters, and when the adjustment module is used to adjust the external parameter information according to the spliced point cloud data, include:

对所述外部参数信息中的至少一个外部参数进行调整，得到调整后的外部参数信息。Adjust at least one external parameter in the external parameter information to obtain adjusted external parameter information.
根据权利要求14所述的标定装置，其特征在于，所述调整模块在用于对所述外部参数信息中的至少一个外部参数进行调整，得到调整后的外部参数信息时，包括：The calibration device according to claim 14, wherein when the adjustment module is used to adjust at least one external parameter in the external parameter information to obtain the adjusted external parameter information, the adjustment module includes:

在对所述外部参数信息进行当前轮调整过程中，从所述多个外部参数中选择当前轮还未调整的外部参数；其中，对外部参数信息进行一轮调整包括对所述外部参数信息中的各个外部参数均进行调整；During the current round of adjustment of the external parameter information, the external parameters that have not been adjusted in the current round are selected from the plurality of external parameters; wherein, the one-round adjustment of the external parameter information includes adjusting the external parameters in the external parameter information. All external parameters are adjusted;

对选择的外部参数的参数值进行当前次调整，得到当前调整后的外部参数信息；Perform the current adjustment on the parameter value of the selected external parameter to obtain the current adjusted external parameter information;

确定基于当前调整后的外部参数信息得到的拼接点云数据中的点的个数与基于当前次调整前的外部参数信息得到的拼接点云数据中的点的个数相比是否变少；其中，当前次调整前的外部参数信息包括选择的外部参数在进行当前次调整之前的参数值；若变少，对选择的外部参数的参数值进行更新后，返回对选择的外部参数进行当前次调整的过程否则，保持选择的外部参数在当前次调整前的参数值，返回从所述多个外部参数中选择当前轮还未调整的外部参数的过程；Determine whether the number of points in the spliced point cloud data obtained based on the currently adjusted external parameter information is less than the number of points in the spliced point cloud data obtained based on the external parameter information before the current adjustment; wherein , the external parameter information before the current adjustment includes the parameter value of the selected external parameter before the current adjustment; if it becomes less, after updating the parameter value of the selected external parameter, return to the current adjustment of the selected external parameter Otherwise, keep the parameter value of the selected external parameter before the current adjustment, and return to the process of selecting the external parameter that has not been adjusted in the current round from the multiple external parameters;

在对所述外部参数信息完成当前轮调整后，判断当前轮最后选择的外部参数的参数值在当前轮调整前后是否发生变化；若发生变化，进行下一轮外部参数信息调整，若未发生变化，确定对外部参数信息的调整结果达到调整后的外部参数信息与调整前的外部参数信息相同的条件，将调整后的外部参数信息作为所述目标外部参数信息。After completing the current round of adjustment of the external parameter information, determine whether the parameter value of the last selected external parameter in the current round has changed before and after the current round of adjustment; if there is a change, perform the next round of external parameter information adjustment. , determining that the adjustment result of the external parameter information reaches the same condition as the external parameter information after adjustment and the external parameter information before adjustment, and using the external parameter information after adjustment as the target external parameter information.
根据权利要求9至15任一所述的标定装置，其特征在于，所述标定装置还包括构图模块，在得到所述目标外部参数信息后，所述构图模块用于：The calibration device according to any one of claims 9 to 15, wherein the calibration device further comprises a composition module, and after obtaining the target external parameter information, the composition module is used for:

获取所述行驶设备行驶过程中，所述组合惯性导航设备的位姿数据、以及所述雷达传感器采集的三维点云数据；acquiring the pose data of the integrated inertial navigation device and the three-dimensional point cloud data collected by the radar sensor during the driving of the traveling device;

基于所述目标外部参数信息和所述组合惯性导航设备的位姿数据，确定所述雷达传感器的位姿数据；Determine the pose data of the radar sensor based on the target external parameter information and the pose data of the combined inertial navigation device;

基于所述雷达传感器在不同时间点的位姿数据、以及在相应时间点采集的三维点云数据，确定在所述三维点云数据中各个点的位置信息；Determine the position information of each point in the three-dimensional point cloud data based on the pose data of the radar sensor at different time points and the three-dimensional point cloud data collected at the corresponding time points;

基于所述三维点云数据中各个点的位置信息，构建所述行驶设备行驶的区域的地图。Based on the position information of each point in the three-dimensional point cloud data, a map of the area where the traveling device travels is constructed.
一种电子设备，包括处理器、存储器和总线，所述存储器存储有所述处理器可执行的机器可读指令，当电子设备运行时，所述处理器与所述存储器之间通过总线通信，所述机器可读指令被所述处理器执行时执行如权利要求1至8任一所述的标定方法。An electronic device, comprising a processor, a memory and a bus, the memory stores machine-readable instructions executable by the processor, when the electronic device is running, the processor and the memory communicate through the bus, The machine-readable instructions, when executed by the processor, perform the calibration method as claimed in any one of claims 1 to 8 .
一种计算机可读存储介质，其上存储有计算机程序，该计算机程序被处理器运行时执行如权利要求1至8任一所述的标定方法。A computer-readable storage medium on which a computer program is stored, the computer program executes the calibration method according to any one of claims 1 to 8 when the computer program is run by a processor.