WO2023142813A1 - Data fusion method and apparatus based on multi-sensor, device, and medium - Google Patents

Abstract

The present application discloses a data fusion method and apparatus based on a multi-sensor, a device, and a medium. The method comprises: acquiring original sensor data respectively collected by at least two sensors; for each sensor, predicting according to the original sensor data collected by the each sensor to obtain predicted sensor data; determining a data difference sequence of the each sensor according to the original sensor data and the predicted sensor data; determining whether the data difference sequence of the each sensor satisfies an anomaly detection condition, and in response to the situation that it is determined that the data difference sequence of the each sensor satisfies the anomaly detection condition, determining a fusion weight of the sensor according to the data difference sequence of the sensor; and fusing, according to the fusion weights of the at least two sensors, the original sensor data respectively collected by the at least two sensors.

Description

基于多传感器的数据融合方法、装置、设备及介质Multi-sensor based data fusion method, device, equipment and medium

本申请要求在2022年1月26日提交中国专利局、申请号为202210092108.2的中国专利申请的优先权，该申请的全部内容通过引用结合在本申请中。This application claims priority to a Chinese patent application with application number 202210092108.2 filed with the China Patent Office on January 26, 2022, the entire contents of which are incorporated herein by reference.

技术领域technical field

本申请实施例涉及智能汽车技术领域，例如涉及一种基于多传感器的数据融合方法、装置、设备及介质。The embodiments of the present application relate to the technical field of smart cars, for example, to a multi-sensor-based data fusion method, device, device, and medium.

背景技术Background technique

近年来，智能汽车己经成为世界车辆工程领域研究的热点和汽车工业增长的新动力。摄像头、毫米波雷达、激光雷达、超声波雷达等传感器是智能汽车的重要感知设备。这些传感器提供的信息对智能汽车感知外部世界以及决定规划和控制方案起着至关重要的作用。In recent years, smart cars have become a research hotspot in the field of vehicle engineering in the world and a new driving force for the growth of the automotive industry. Sensors such as cameras, millimeter-wave radars, lidars, and ultrasonic radars are important sensing devices for smart cars. The information provided by these sensors plays a vital role for smart cars to perceive the external world and decide on planning and control schemes.

由于智能网联车辆在环境感知的过程中需要通过多种传感器的协同工作，不同传感器之间的数据融合一直是网联车辆的研究热点，相关技术中的数据融合过程中，大多是根据传感器的等级确定该传感器采集的数据在融合过程中的使用权重，而传感器等级的划分是根据传感器的类别进行提前预设的，因此，对应的传感器等级是固定不变的，导致传感器数据的使用权重在多种场景下都不会发生变化，在某些场景下，容易出现数据融合不准确，导致环境感知出现偏差的现象，最终影响控制器决策的正确性，亟需改进。Since intelligent networked vehicles need to work collaboratively through multiple sensors in the process of environment perception, data fusion between different sensors has always been a research hotspot for networked vehicles. In the process of data fusion in related technologies, most of them are based on sensor The level determines the use weight of the data collected by the sensor in the fusion process, and the division of the sensor level is preset in advance according to the sensor category. Therefore, the corresponding sensor level is fixed, resulting in the use weight of the sensor data. It will not change in a variety of scenarios. In some scenarios, inaccurate data fusion is prone to occur, leading to deviations in environmental perception, which ultimately affects the correctness of the controller's decision-making, and urgently needs to be improved.

发明内容Contents of the invention

本申请提供一种基于多传感器的数据融合方法、装置、设备及介质，以根据实时数据调节融合权重，提高融合结果的准确性。The present application provides a data fusion method, device, equipment and medium based on multi-sensors, so as to adjust fusion weights according to real-time data and improve the accuracy of fusion results.

第一方面，本申请实施例提供了一种基于多传感器的数据融合方法，所述方法包括：In the first aspect, the embodiment of the present application provides a multi-sensor-based data fusion method, the method comprising:

获取至少两种传感器分别采集的原传感器数据；Obtaining raw sensor data collected by at least two sensors;

针对每种传感器，根据每种传感器采集的原传感器数据进行预测，得到预测传感器数据；For each sensor, predict according to the original sensor data collected by each sensor, and obtain the predicted sensor data;

根据所述原传感器数据和所述预测传感器数据，确定所述每种传感器的数据差值序列；According to the original sensor data and the predicted sensor data, determine the data difference sequence of each sensor;

确定所述每种传感器的数据差值序列是否满足异常检测条件，响应于确定所述每种传感器的数据差值序列满足异常检测条件，根据所述每种传感器的数据差值序列确定所述每种传感器的融合权重；Determining whether the data difference sequence of each sensor satisfies the abnormality detection condition, in response to determining that the data difference sequence of each sensor satisfies the abnormality detection condition, determining the each sensor according to the data difference sequence of each sensor Fusion weights of various sensors;

根据至少两种传感器的融合权重，对至少两种传感器分别采集的原传感器数据进行融合。According to the fusion weights of the at least two sensors, the original sensor data respectively collected by the at least two sensors are fused.

第二方面，本申请实施例还提供了一种基于多传感器的数据融合装置，所 述装置包括：In the second aspect, the embodiment of the present application also provides a multi-sensor-based data fusion device, the device comprising:

数据获取模块，设置为获取至少两种传感器分别采集的原传感器数据；The data acquisition module is configured to acquire the original sensor data collected by at least two sensors;

数据预测模块，设置为针对每种传感器，根据每种传感器采集的原传感器数据进行预测，得到预测传感器数据；The data prediction module is configured to perform prediction according to the original sensor data collected by each sensor to obtain predicted sensor data for each sensor;

差值确定模块，设置为根据所述每种原传感器数据和所述预测传感器数据，确定所述每种传感器的数据差值序列；The difference determination module is configured to determine the data difference sequence of each sensor according to the original sensor data and the predicted sensor data;

权重确定模块，设置为确定所述每种传感器的数据差值序列是否满足异常检测条件，响应于确定所述每种传感器的数据差值序列满足异常检测条件，根据所述每种传感器的数据差值序列确定所述每种传感器的融合权重；A weight determination module, configured to determine whether the data difference sequence of each sensor satisfies the abnormality detection condition, in response to determining that the data difference sequence of each sensor satisfies the abnormality detection condition, according to the data difference sequence of each sensor a sequence of values determining fusion weights for each of the sensors;

数据融合模块，设置为根据所述至少两种传感器的融合权重，对所述至少两种传感器分别采集的原传感器数据进行融合。The data fusion module is configured to fuse the original sensor data respectively collected by the at least two sensors according to the fusion weights of the at least two sensors.

第三方面，本申请实施例还提供了一种电子设备，该电子设备包括：In a third aspect, the embodiment of the present application also provides an electronic device, the electronic device includes:

一个或多个处理器；one or more processors;

存储装置，设置为存储一个或多个程序，storage means configured to store one or more programs,

当所述一个或多个程序被所述一个或多个处理器执行，使得所述一个或多个处理器实现本申请任意实施例所述的基于多传感器的数据融合方法。When the one or more programs are executed by the one or more processors, the one or more processors implement the multi-sensor-based data fusion method described in any embodiment of the present application.

第四方面，本申请实施例还提供了一种计算机可读存储介质，其上存储有计算机程序，该程序被处理器执行时实现如本申请任意实施例所述的基于多传感器的数据融合方法。In the fourth aspect, the embodiment of the present application also provides a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, the multi-sensor-based data fusion method as described in any embodiment of the present application is implemented. .

附图说明Description of drawings

图1是本申请一实施例提供的一种基于多传感器的数据融合方法的流程图；FIG. 1 is a flowchart of a multi-sensor-based data fusion method provided by an embodiment of the present application;

图2是本申请另一实施例提供的一种基于多传感器的数据融合方法的流程图；FIG. 2 is a flowchart of a multi-sensor-based data fusion method provided by another embodiment of the present application;

图3是本申请另一实施例提供的一种基于多传感器的数据融合方法的流程图；FIG. 3 is a flow chart of a multi-sensor-based data fusion method provided by another embodiment of the present application;

图4是本申请一实施例提供的一种基于多传感器的数据融合装置结构框图；FIG. 4 is a structural block diagram of a multi-sensor-based data fusion device provided by an embodiment of the present application;

图5是本申请一实施例提供的一种电子设备的结构示意图。Fig. 5 is a schematic structural diagram of an electronic device provided by an embodiment of the present application.

具体实施方式Detailed ways

图1为本申请一实施例提供的一种基于多传感器的数据融合方法的流程图，本实施例可适用于对多传感器的数据进行融合的情况，例如适用于智能驾驶***中采用多传感器感知确保感知的全面性。该方法可以由本申请实施例提供的基于多传感器的数据融合装置来执行，该装置可以采用软件和/或硬件的方式实现，并可集成在电子设备上。Figure 1 is a flowchart of a multi-sensor-based data fusion method provided by an embodiment of the present application. This embodiment can be applied to the situation of fusing multi-sensor data, for example, it is applicable to the use of multi-sensor perception in intelligent driving systems. Ensure comprehensiveness of perception. The method can be executed by the multi-sensor-based data fusion device provided in the embodiment of the present application. The device can be implemented in the form of software and/or hardware, and can be integrated on the electronic device.

例如，如图1所示，本申请实施例提供的基于多传感器的数据融合方法，可以包括如下步骤：For example, as shown in Figure 1, the multi-sensor-based data fusion method provided in the embodiment of the present application may include the following steps:

S110、获取至少两种传感器分别采集的原传感器数据。S110. Acquire raw sensor data respectively collected by at least two sensors.

其中，传感器主要用于感知周围环境，设置在车端，可以包括：毫米波雷达、激光雷达、单\双目摄像头以及卫星导航等。原传感器数据，可以理解为，传感器采集的原始数据。Among them, the sensor is mainly used to perceive the surrounding environment, and it is installed on the vehicle end, which can include: millimeter-wave radar, lidar, single/binocular camera, and satellite navigation. The original sensor data can be understood as the original data collected by the sensor.

在车辆行驶过程中，可以通过安装在车上的各式各样的传感器来感应周围的环境，收集数据，生成原传感器数据，以便后续根据原传感器数据进行静态、动态物体的辨识、侦测与追踪，进行***的运算与分析，从而预先让驾驶者察觉到可能发生的危险，有效增加汽车驾驶的舒适性和安全性。During the driving process of the vehicle, various sensors installed on the vehicle can be used to sense the surrounding environment, collect data, and generate original sensor data for subsequent identification, detection and monitoring of static and dynamic objects based on the original sensor data. Tracking, calculation and analysis of the system, so that the driver can be aware of possible dangers in advance, effectively increasing the comfort and safety of driving.

例如，可以根据至少两种传感器分别采集的原传感器数据，确定至少两个传感器的原传感器采集数据中是否存在关联目标，若存在关联目标，则将关联的原传感器数据按融合权重进行数据融合。例如，可以设置一个融合关联门限，若两个传感器中的目标在融合关联门限的数值范围内，则确定两者关联。For example, based on the original sensor data collected by at least two sensors, it may be determined whether there is an associated target in the original sensor data collected by at least two sensors, and if there is an associated object, the associated original sensor data is fused according to the fusion weight. For example, a fusion association threshold may be set, and if the targets in the two sensors are within the value range of the fusion association threshold, the two sensors are determined to be associated.

由于多传感器的检测精度会受到车辆行驶环境的影响，而车辆行驶环境又是动态变化的，因此，在进行多传感器数据融合时，如果采用固定不变的多个传感器的融合优先级或融合权重来进行数据融合，将无法保证待检测目标的定位精度。Since the detection accuracy of multi-sensors will be affected by the driving environment of the vehicle, and the driving environment of the vehicle is dynamically changing, when performing multi-sensor data fusion, if the fusion priority or fusion weight of multiple sensors is fixed For data fusion, the positioning accuracy of the target to be detected will not be guaranteed.

另外，需要说明的是，传感器的运行状态标志位为正常、性能受限和故障三种状态，在三种不同状态下，传感器采集的原传感器数据的准确性存在差异。考虑到传感器运行状态对数据融合的影响，可以在传感器性能波动时对传感器的融合权重进行调整，以增强数据融合效果。In addition, it should be noted that the operating state flags of the sensor are three states: normal, performance limited, and failure. In the three different states, the accuracy of the original sensor data collected by the sensor is different. Considering the impact of sensor operation status on data fusion, the sensor fusion weight can be adjusted when the sensor performance fluctuates to enhance the data fusion effect.

S120、针对每种传感器，根据每种传感器采集的原传感器数据进行预测，得到预测传感器数据。S120. For each sensor, perform prediction according to the original sensor data collected by each sensor, to obtain predicted sensor data.

本实施例中，每种传感器的融合权重受该传感器的运行状态影响，与其他传感器没有直接关联。因此，每种传感器依靠该传感器自身采集的原传感器数据和运行状态，确定自身的融合权重。In this embodiment, the fusion weight of each sensor is affected by the operating state of the sensor, and is not directly related to other sensors. Therefore, each sensor determines its own fusion weight depending on the original sensor data and operating status collected by the sensor itself.

在一个示例实施方式中，根据每种传感器采集的原传感器数据进行预测，可以是根据传感器在t-1时刻采集的传感器数据，预测其在t时刻的传感器数据。在另一个示例实施方式中，根据每种传感器采集的原传感器数据进行预测，还可以是根据传感器在t-1、t-2、…、t-k时刻的采集的传感器数据，预测其在t时刻的传感器数据，得到预测传感器数据。In an example implementation, the prediction is performed based on the original sensor data collected by each sensor, which may be to predict the sensor data at time t based on the sensor data collected by the sensor at time t-1. In another exemplary embodiment, the prediction is made based on the original sensor data collected by each sensor, or the sensor data collected by the sensor at time t-1, t-2, ..., t-k can be used to predict its sensor data, to get predicted sensor data.

例如，预测传感器数据可以由原传感器数据通过卡尔曼滤波确定。For example, predicted sensor data may be determined from raw sensor data by Kalman filtering.

S130、根据原传感器数据和预测传感器数据，确定每种传感器的数据差值序列。S130. Determine a data difference sequence of each sensor according to the original sensor data and the predicted sensor data.

其中，数据差值序列是传感器在同一时刻的预测传感器数据和原传感器数据的差值。例如，若t时刻的原传感器数据为A _t,t时刻的预测传感器数据为B _t，则差值为A _t-B _t。 Wherein, the data difference sequence is the difference between the predicted sensor data and the original sensor data of the sensor at the same time. For example, if the original sensor data at time t is A _t and the predicted sensor data at time t is B _t , then the difference is A _t −B _t .

将每种传感器的原传感器数据和预测传感器数据作差，即可确定该传感器的数据差值序列。By making a difference between the original sensor data and the predicted sensor data of each sensor, the data difference sequence of the sensor can be determined.

S140、确定每种传感器的数据差值序列是否满足异常检测条件，响应于确定每种传感器的数据差值序列满足异常检测条件，根据每种传感器的数据差值 序列确定每种传感器的融合权重。S140. Determine whether the data difference sequence of each sensor satisfies the anomaly detection condition, and determine the fusion weight of each sensor according to the data difference sequence of each sensor in response to determining that the data difference sequence of each sensor satisfies the anomaly detection condition.

其中，满足异常检测条件是对传感器的融合权重进行调整的前提。如果传感器的差值序列满足异常检测条件，则根据传感器的数据差值序列确定传感器的融合权重，如果传感器的差值序列不满足异常检测条件，以该传感器的基础权重作为融合权重。Among them, satisfying the anomaly detection condition is the prerequisite for adjusting the fusion weight of the sensor. If the difference sequence of the sensor satisfies the anomaly detection condition, the fusion weight of the sensor is determined according to the difference sequence of the sensor data; if the difference sequence of the sensor does not meet the anomaly detection condition, the basic weight of the sensor is used as the fusion weight.

例如，异常检测条件，可以为传感器的数据差值序列中存在大于异常检测阈值的数据。确定传感器的数据差值序列是否满足异常检测条件，包括：确定传感器的数据差值序列中是否存在大于异常检测阈值的数据。其中，异常检测阈值，可以由传感器运行状态正常时从原传感器数据中筛选的能够较长时间稳定跟踪目标的数据段确定。For example, the anomaly detection condition may be that there is data greater than the anomaly detection threshold in the data difference sequence of the sensor. Determining whether the data difference sequence of the sensor satisfies an abnormality detection condition includes: determining whether there is data greater than an abnormality detection threshold in the sensor data difference sequence. Among them, the abnormal detection threshold can be determined by the data segment that can stably track the target for a long time and is screened from the original sensor data when the sensor is running normally.

例如，可以从原传感器数据中筛选出每种传感器标志位正常且状态稳定时的数据，作为该传感器的目标传感器数据；根据目标传感器数据，确定目标预测数据；根据目标预测数据和目标传感器数据，确定目标差值序列；对目标差值序列进行分段，计算每段数据的标准差，形成目标标准差序列；根据目标标准差序列，确定传感器的所述异常检测阈值。For example, the data when each sensor flag is normal and in a stable state can be screened out from the original sensor data as the target sensor data of the sensor; according to the target sensor data, the target prediction data is determined; according to the target prediction data and the target sensor data, determining the target difference sequence; segmenting the target difference sequence, calculating the standard deviation of each piece of data to form a target standard deviation sequence; determining the abnormality detection threshold of the sensor according to the target standard deviation sequence.

S150、根据至少两种传感器的融合权重，对至少两种传感器分别采集的原传感器数据进行融合。S150. According to the fusion weights of the at least two sensors, fuse the original sensor data respectively collected by the at least two sensors.

其中，传感器的融合权重可以是该传感器的基础权重，也可以是根据传感器的数据差值序列确定权重。基础权重值，由传感器的类型和融合属性确定。Wherein, the fusion weight of the sensor may be the basic weight of the sensor, or the weight determined according to the data difference sequence of the sensor. The base weight value, determined by the sensor type and fusion properties.

例如，根据至少两种传感器的融合权重，对至少两种传感器分别采集的原传感器数据进行融合，可以是如果传感器的差值序列满足异常检测条件，则根据传感器的数据差值序列确定传感器的融合权重，如果传感器的差值序列不满足异常检测条件，以该传感器的基础权重作为融合权重，进行数据融合。For example, according to the fusion weights of at least two sensors, the original sensor data collected by at least two sensors are fused, which may be that if the sensor difference sequence satisfies the abnormal detection condition, then determine the fusion of the sensor according to the sensor data difference sequence Weight, if the difference sequence of the sensor does not meet the anomaly detection condition, the basic weight of the sensor is used as the fusion weight for data fusion.

本实施例的技术方案，通过获取至少两种传感器分别采集的原传感器数据；针对每种传感器，根据每种传感器采集的原传感器数据进行预测，得到预测传感器数据；根据原传感器数据和预测传感器数据，确定传感器的数据差值序列；确定传感器的数据差值序列是否满足异常检测条件，响应于确定传感器的数据差值序列满足异常检测条件，根据传感器的数据差值序列确定传感器的融合权重；根据至少两种传感器的融合权重，对至少两种传感器分别采集的原传感器数据进行融合。通过本实施例的技术方案，可以避免相关技术中多个传感器的融合权重固定导致的环境感知偏差情况，根据实时数据调节融合权重，提高融合结果的准确性，进而有利于提高控制器决策的正确性，为多传感器的数据融合提供了一种新思路。In the technical solution of this embodiment, by obtaining the original sensor data collected by at least two sensors respectively; for each sensor, prediction is made according to the original sensor data collected by each sensor to obtain the predicted sensor data; according to the original sensor data and the predicted sensor data , determine the data difference sequence of the sensor; determine whether the data difference sequence of the sensor satisfies the anomaly detection condition, in response to determining that the sensor data difference sequence satisfies the anomaly detection condition, determine the fusion weight of the sensor according to the sensor data difference sequence; The fusion weights of the at least two sensors are used to fuse the original sensor data respectively collected by the at least two sensors. Through the technical solution of this embodiment, the environmental perception deviation caused by the fixed fusion weights of multiple sensors in the related art can be avoided, the fusion weights can be adjusted according to real-time data, and the accuracy of the fusion results can be improved, which in turn helps to improve the correctness of the controller's decision-making It provides a new idea for multi-sensor data fusion.

图2为本申请另一实施例提供的一种基于多传感器的数据融合方法的流程图，该方法在上述实施例的基础上进一步的细化，给出了确定异常检测阈值的具体情况介绍。FIG. 2 is a flow chart of a multi-sensor based data fusion method provided by another embodiment of the present application. The method is further refined on the basis of the above embodiment, and the specific situation of determining the abnormality detection threshold is given.

例如，如图2所示，该方法包括：For example, as shown in Figure 2, the method includes:

S210、获取至少两种传感器分别采集的原传感器数据。S210. Acquire raw sensor data respectively collected by at least two sensors.

S220、从原传感器数据中筛选出每种传感器标志位正常且状态稳定时的数据，作为该传感器的目标传感器数据。S220. Select the data when the flag bit of each sensor is normal and the state is stable from the original sensor data, and use it as the target sensor data of the sensor.

其中，传感器标志位可以根据传感器的运行状态简单分为三类：正常、性能受限和故障。可以理解的是，为了保证确定的异常检测阈值的有效性和准确性，从原传感器数据中筛选出每种传感器标志位正常的数据。Among them, the sensor flags can be simply divided into three categories according to the operating status of the sensor: normal, performance limited and failure. It can be understood that, in order to ensure the validity and accuracy of the determined abnormality detection threshold, the data with normal flag bits of each sensor is screened out from the original sensor data.

另外，考虑到并非每一帧原传感器数据都有利于后续目标跟踪识别，可以从原传感器数据中筛选出能较长时间稳定跟踪的原传感器数据，即状态稳定时的数据。In addition, considering that not every frame of original sensor data is conducive to subsequent target tracking and identification, the original sensor data that can be tracked stably for a long time can be screened out from the original sensor data, that is, the data when the state is stable.

例如，可以从原传感器数据中筛选出每种传感器标志位正常且状态稳定时的数据，作为该传感器的目标传感器数据。For example, the data when the flag bit of each sensor is normal and in a stable state can be screened out from the original sensor data as the target sensor data of the sensor.

S230、根据目标传感器数据，确定目标预测数据。S230. Determine target prediction data according to the target sensor data.

需要说明的是，根据目标传感器数据确定目标预测数据的方法，与上述实施例中根据每种传感器采集的原传感器数据进行预测，得到预测传感器数据的方法一致。It should be noted that the method of determining the target prediction data based on the target sensor data is consistent with the method of obtaining the predicted sensor data by performing prediction based on the original sensor data collected by each sensor in the above embodiment.

若根据每种传感器采集的原传感器数据进行预测，得到预测传感器数据时，根据传感器在t-1时刻采集的传感器数据，预测其在t时刻的传感器数据。则根据目标传感器数据，确定目标预测数据，例如，根据t-1时刻的目标传感器数据，确定t时刻的目标预测数据。If the prediction is made based on the original sensor data collected by each sensor, when the predicted sensor data is obtained, the sensor data at time t is predicted based on the sensor data collected by the sensor at time t-1. Then, according to the target sensor data, the target prediction data is determined, for example, according to the target sensor data at time t-1, the target prediction data at time t is determined.

若根据每种传感器采集的原传感器数据进行预测，得到预测传感器数据时，根据传感器在t-1、t-2、…、t-k时刻的采集的传感器数据，预测其在t时刻的传感器数据，得到预测传感器数据。则根据目标传感器数据，确定目标预测数据，例如，根据t-1、t-2、…、t-k时刻的目标传感器数据，确定t时刻的目标预测数据。If the prediction is made based on the original sensor data collected by each sensor, when the predicted sensor data is obtained, the sensor data collected by the sensor at time t-1, t-2, ..., t-k are used to predict the sensor data at time t, and we get Predict sensor data. Then, according to the target sensor data, the target prediction data is determined, for example, according to the target sensor data at time t-1, t-2, ..., t-k, the target prediction data at time t is determined.

例如，目标预测数据可以由目标传感器数据通过卡尔曼滤波确定。For example, target prediction data can be determined from target sensor data through Kalman filtering.

S240、根据目标预测数据和目标传感器数据，确定目标差值序列。S240. Determine a target difference sequence according to the target prediction data and the target sensor data.

将每种传感器的目标传感器数据和目标预测数据作差，即可确定该传感器的目标差值序列。By making a difference between the target sensor data and the target prediction data of each sensor, the target difference sequence of the sensor can be determined.

S250、对目标差值序列进行分段，计算每段数据的标准差，形成目标标准差序列。S250. Segment the target difference sequence, calculate the standard deviation of each segment of data, and form a target standard deviation sequence.

由于每一帧目标传感器数据对应的目标差值序列的数据量较大，可以将目标差值序列以相同目标数分为T段。计算每段目标差值序列的标准差θ _i，并形成长度为T的目标标准差序列α＝[θ ₁,θ ₂,…,θ _T]。 Since the target difference sequence corresponding to each frame of target sensor data has a large amount of data, the target difference sequence can be divided into T segments with the same number of targets. Calculate the standard deviation θ _i of each target difference sequence, and form a target standard deviation sequence α=[θ ₁ ,θ ₂ ,…,θ _T ] with length T.

S260、根据目标标准差序列，确定传感器的异常检测阈值。S260. Determine an abnormality detection threshold of the sensor according to the target standard deviation sequence.

其中，异常检测阈值用于判断是否需要对传感器的融合权重进行调整。如果传感器的数据差值序列中存在大于异常检测阈值的数据，则根据传感器的数据差值序列确定传感器的融合权重，如果传感器的数据差值序列中存在小于或等于异常检测阈值的数据，以该传感器的基础权重作为融合权重。Among them, the anomaly detection threshold is used to judge whether it is necessary to adjust the fusion weight of the sensor. If there is data greater than the anomaly detection threshold in the data difference sequence of the sensor, the fusion weight of the sensor is determined according to the data difference sequence of the sensor, and if there is data less than or equal to the abnormal detection threshold in the sensor data difference sequence, the The base weights of the sensors are used as fusion weights.

确定目标标准差序列α后，可以分析目标标准差序列α的分布情况。结合目标标准差序列α的分布情况，按照预设的发生概率确定该传感器对应的异常检测 阈值。After determining the target standard deviation sequence α, the distribution of the target standard deviation sequence α can be analyzed. Combined with the distribution of the target standard deviation sequence α, the abnormal detection threshold corresponding to the sensor is determined according to the preset probability of occurrence.

例如，根据目标标准差序列，确定传感器的异常检测阈值，可以包括：根据目标标准差序列，确定目标标准差序列的均值和标准差；基于目标标准差序列的均值和标准差，以发生概率小于50％的原则，确定传感器的异常检测阈值。For example, determining the abnormality detection threshold of the sensor according to the target standard deviation sequence may include: determining the mean value and standard deviation of the target standard deviation sequence according to the target standard deviation sequence; The 50% principle determines the anomaly detection threshold of the sensor.

本实施例的技术方案，给出了确定异常检测阈值的具体情况介绍，通过从原传感器数据中筛选出每种传感器标志位正常且状态稳定时的数据，作为该传感器的目标传感器数据；根据目标传感器数据，确定目标预测数据；根据目标预测数据和目标传感器数据，确定目标差值序列；对目标差值序列进行分段，计算每段数据的标准差，形成目标标准差序列；根据目标标准差序列，确定传感器的异常检测阈值。通过从原传感器数据中筛选传感器标志位正常且状态稳定的数据，并据此确定异常检测阈值，避免了相关技术中多个传感器的融合权重固定导致的环境感知偏差情况，通过异常检测阈值确定多个传感器的运行状态是否异常，并结合实时数据调节融合权重，提高融合结果的准确性。The technical scheme of this embodiment provides the specific situation introduction of determining the abnormality detection threshold, by screening out the data when each sensor flag is normal and in a stable state from the original sensor data, as the target sensor data of the sensor; according to the target Determine the target prediction data from the sensor data; determine the target difference sequence according to the target prediction data and the target sensor data; segment the target difference sequence, calculate the standard deviation of each segment of data, and form the target standard deviation sequence; according to the target standard deviation sequence, which determines the sensor's anomaly detection threshold. By screening the data with normal sensor flags and stable states from the original sensor data, and determining the abnormality detection threshold based on this, it avoids the environment perception deviation caused by the fixed fusion weight of multiple sensors in the related technology, and determines the abnormality detection threshold through the abnormality detection threshold. Whether the operating status of each sensor is abnormal, and adjust the fusion weight in combination with real-time data to improve the accuracy of the fusion result.

图3为本申请另一实施例提供的一种基于多传感器的数据融合方法的流程图，该方法在上述实施例的基础上进一步的细化，给出了确定传感器的融合权重的具体情况介绍。Fig. 3 is a flow chart of a multi-sensor based data fusion method provided by another embodiment of the present application. This method is further refined on the basis of the above embodiment, and the specific situation of determining the fusion weight of the sensor is given. .

例如，如图3所示，该方法包括：For example, as shown in Figure 3, the method includes:

S310、获取至少两种传感器分别采集的原传感器数据。S310. Acquire raw sensor data respectively collected by at least two sensors.

S320、针对每种传感器，根据每种传感器采集的原传感器数据进行预测，得到预测传感器数据。S320. For each sensor, perform prediction according to the original sensor data collected by each sensor, to obtain predicted sensor data.

S330、根据原传感器数据和预测传感器数据，确定传感器的数据差值序列。S330. Determine a sensor data difference sequence according to the original sensor data and the predicted sensor data.

S340、确定传感器的数据差值序列是否满足异常检测条件，响应于确定传感器的数据差值序列满足异常检测条件，执行S350；响应于确定传感器的数据差值序列不满足异常检测条件，执行S370。S340. Determine whether the data difference sequence of the sensor satisfies the abnormality detection condition, and execute S350 in response to determining that the sensor data difference sequence meets the abnormality detection condition; and execute S370 in response to determining that the sensor data difference sequence does not meet the abnormality detection condition.

S350、对传感器的数据差值序列和目标差值序列进行一致性检测，确定一致性检测值。S350. Perform consistency detection on the sensor data difference sequence and the target difference sequence, and determine a consistency detection value.

其中，一致性检测值，可以表现传感器的数据差值序列和目标差值序列的分布相似程度。Among them, the consistency detection value can represent the distribution similarity between the sensor data difference sequence and the target difference sequence.

对传感器的数据差值序列和目标差值序列进行一致性检测，可以理解为，对传感器的数据差值序列和目标差值序列的分布进行分析比较。示例性的，可以对传感器的数据差值序列和目标差值序列进行KS(Kolmogorov-Smirnov)一致性检测，得到一致性检测值p。The consistency detection of the sensor data difference sequence and the target difference sequence can be understood as analyzing and comparing the distributions of the sensor data difference sequence and the target difference sequence. Exemplarily, KS (Kolmogorov-Smirnov) consistency detection may be performed on the sensor data difference sequence and the target difference sequence to obtain the consistency detection value p.

S360、根据一致性检测值，确定传感器的融合权重。S360. Determine fusion weights of the sensors according to the consistency detection value.

例如，根据一致性检测值，确定传感器的融合权重，可以包括：响应于确定一致性检测值大于一致性检测阈值，根据传感器的基础权重值和一致性检测值确定传感器的融合权重；响应于确定一致性检测值小于或等于一致性检测阈值，根据传感器的基础权重值确定传感器的融合权重。For example, determining the fusion weight of the sensor according to the consistency detection value may include: determining the fusion weight of the sensor according to the basic weight value and the consistency detection value of the sensor in response to determining that the consistency detection value is greater than the consistency detection threshold; The consistency detection value is less than or equal to the consistency detection threshold, and the fusion weight of the sensor is determined according to the basic weight value of the sensor.

示例性的，若一致性检测值p≤0.05，则融合权重＝0.5*基础权重值；若p＞ 0.05，融合权重＝p*基础权重值。其中，基础权重值由该传感器的类型和融合属性决定。需要说明的是，实际每个传感器性能不一样，p值与融合权重的对应关系也是不一样的，需要根据实际情况进行调整。Exemplarily, if the consistency detection value p≤0.05, the fusion weight=0.5*basic weight value; if p>0.05, the fusion weight=p*basic weight value. Among them, the basic weight value is determined by the type and fusion properties of the sensor. It should be noted that the actual performance of each sensor is different, and the corresponding relationship between the p value and the fusion weight is also different, which needs to be adjusted according to the actual situation.

S370、根据至少两种传感器的融合权重，对至少两种传感器分别采集的原传感器数据进行融合。S370. According to the fusion weights of the at least two sensors, fuse the original sensor data respectively collected by the at least two sensors.

本实施例的技术方案，给出了确定传感器的融合权重的具体情况介绍，通过在确定传感器的数据差值序列是否满足异常检测条件的情况下，对传感器的数据差值序列和目标差值序列进行一致性检测，确定一致性检测值，根据一致性检测值，确定传感器的融合权重，可以根据传感器的运行状态实时调节融合权重，提高融合结果的准确性，进而有利于提高控制器决策的正确性。The technical solution of this embodiment provides an introduction to the specific situation of determining the fusion weight of the sensor. By determining whether the data difference sequence of the sensor meets the abnormal detection condition, the data difference sequence of the sensor and the target difference sequence Carry out consistency detection, determine the consistency detection value, determine the fusion weight of the sensor according to the consistency detection value, adjust the fusion weight in real time according to the operating status of the sensor, improve the accuracy of the fusion result, and then help improve the correctness of the controller's decision-making sex.

图4是本申请实施例所提供的一种基于多传感器的数据融合装置的结构示意图，该装置适用于执行本申请实施例提供的基于多传感器的数据融合方法，可以根据实时数据调节融合权重，提高融合结果的准确性。如图4所示，该装置包括数据获取模块410、数据预测模块420、差值确定模块430、权重确定模块440和数据融合模块450。FIG. 4 is a schematic structural diagram of a multi-sensor-based data fusion device provided in an embodiment of the present application. The device is suitable for implementing the multi-sensor-based data fusion method provided in the embodiment of the present application, and can adjust fusion weights according to real-time data. Improve the accuracy of fusion results. As shown in FIG. 4 , the device includes a data acquisition module 410 , a data prediction module 420 , a difference determination module 430 , a weight determination module 440 and a data fusion module 450 .

其中，数据获取模块410，设置为获取至少两种传感器分别采集的原传感器数据；Wherein, the data acquisition module 410 is configured to acquire the original sensor data collected by at least two sensors;

数据预测模块420，设置为针对每种传感器，根据每种传感器采集的原传感器数据进行预测，得到预测传感器数据；The data prediction module 420 is configured to perform prediction according to the original sensor data collected by each sensor for each sensor, and obtain predicted sensor data;

差值确定模块430，设置为根据原传感器数据和预测传感器数据，确定传感器的数据差值序列；The difference determination module 430 is configured to determine the sensor data difference sequence according to the original sensor data and the predicted sensor data;

权重确定模块440，设置为确定传感器的数据差值序列是否满足异常检测条件，响应于确定传感器的数据差值序列满足异常检测条件，根据传感器的数据差值序列确定传感器的融合权重；The weight determination module 440 is configured to determine whether the data difference sequence of the sensor satisfies the abnormal detection condition, and determines the fusion weight of the sensor according to the sensor data difference sequence in response to determining that the sensor data difference sequence meets the abnormal detection condition;

数据融合模块450，设置为根据至少两种传感器的融合权重，对至少两种传感器分别采集的原传感器数据进行融合。The data fusion module 450 is configured to fuse the original sensor data respectively collected by the at least two sensors according to the fusion weights of the at least two sensors.

本实施例的技术方案，通过获取至少两种传感器分别采集的原传感器数据；针对每种传感器，根据每种传感器采集的原传感器数据进行预测，得到预测传感器数据；根据原传感器数据和预测传感器数据，确定传感器的数据差值序列；确定传感器的数据差值序列是否满足异常检测条件，响应于确定传感器的数据差值序列满足异常检测条件，根据传感器的数据差值序列确定传感器的融合权重；根据至少两种传感器的融合权重，对至少两种传感器分别采集的原传感器数据进行融合，可以避免相关技术中多个传感器的融合权重固定导致的环境感知偏差情况，根据实时数据调节融合权重，提高融合结果的准确性，进而有利于提高控制器决策的正确性，为多传感器的数据融合提供了一种新思路。In the technical solution of this embodiment, by obtaining the original sensor data collected by at least two sensors respectively; for each sensor, prediction is made according to the original sensor data collected by each sensor to obtain the predicted sensor data; according to the original sensor data and the predicted sensor data , determine the data difference sequence of the sensor; determine whether the data difference sequence of the sensor satisfies the anomaly detection condition, in response to determining that the sensor data difference sequence satisfies the anomaly detection condition, determine the fusion weight of the sensor according to the sensor data difference sequence; The fusion weights of at least two sensors, and the fusion of the original sensor data collected by at least two sensors, can avoid the environmental perception deviation caused by the fixed fusion weights of multiple sensors in related technologies, adjust the fusion weights according to real-time data, and improve fusion. The accuracy of the results will help improve the correctness of the controller's decision-making, and provide a new idea for multi-sensor data fusion.

例如，权重确定模块440，设置为确定传感器的数据差值序列中是否存在大于异常检测阈值的数据。For example, the weight determination module 440 is configured to determine whether there is data greater than the abnormality detection threshold in the sensor data difference sequence.

例如，装置还包括：For example, the device also includes:

数据筛选模块，设置为从原传感器数据中筛选出每种传感器标志位正常且状态稳定时的数据，作为该传感器的目标传感器数据；The data screening module is configured to filter out the data when the flag bit of each sensor is normal and in a stable state from the original sensor data, as the target sensor data of the sensor;

目标预测模块，设置为根据目标传感器数据，确定目标预测数据；The target prediction module is configured to determine the target prediction data according to the target sensor data;

目标差值确定模块，设置为根据目标预测数据和目标传感器数据，确定目标差值序列；The target difference determination module is configured to determine the target difference sequence according to the target prediction data and the target sensor data;

序列分段模块，设置为对目标差值序列进行分段，计算每段数据的标准差，形成目标标准差序列；The sequence segmentation module is configured to segment the target difference sequence, calculate the standard deviation of each segment of data, and form the target standard deviation sequence;

阈值确定模块，设置为根据目标标准差序列，确定传感器的异常检测阈值。The threshold determination module is configured to determine the anomaly detection threshold of the sensor according to the target standard deviation sequence.

例如，上述阈值确定模块包括：分布确定单元和阈值确定单元。For example, the above-mentioned threshold determination module includes: a distribution determination unit and a threshold determination unit.

其中，分布确定单元，设置为根据目标标准差序列，确定目标标准差序列的均值和标准差；Wherein, the distribution determination unit is set to determine the mean value and standard deviation of the target standard deviation sequence according to the target standard deviation sequence;

阈值确定单元，设置为基于目标标准差序列的均值和标准差，以发生概率小于50％的原则，确定传感器的异常检测阈值。The threshold determination unit is configured to determine the abnormality detection threshold of the sensor based on the principle that the occurrence probability is less than 50% based on the mean value and standard deviation of the target standard deviation sequence.

例如，上述权重确定模块440包括：一致性检测单元和权重确定单元。For example, the above-mentioned weight determination module 440 includes: a consistency detection unit and a weight determination unit.

其中，一致性检测单元，设置为对传感器的数据差值序列和目标差值序列进行一致性检测，确定一致性检测值；Wherein, the consistency detection unit is configured to perform consistency detection on the sensor data difference sequence and the target difference sequence, and determine the consistency detection value;

权重确定单元，设置为根据一致性检测值，确定传感器的融合权重。The weight determination unit is configured to determine the fusion weight of the sensor according to the consistency detection value.

例如，上述权重确定单元，设置为响应于确定一致性检测值大于一致性检测阈值，根据传感器的基础权重值和一致性检测值确定传感器的融合权重；响应于确定一致性检测值小于或等于一致性检测阈值，根据传感器的基础权重值确定传感器的融合权重；For example, the above-mentioned weight determination unit is configured to determine the fusion weight of the sensor according to the basic weight value and the consistency detection value of the sensor in response to determining that the consistency detection value is greater than the consistency detection threshold; in response to determining that the consistency detection value is less than or equal to the consistency Determine the fusion weight of the sensor according to the basic weight value of the sensor;

例如，装置还包括：基础权重确定模块，设置为根据传感器的类型和融合属性，确定传感器的基础权重值。For example, the device further includes: a basic weight determination module configured to determine the basic weight value of the sensor according to the type of the sensor and the fusion property.

本申请实施例所提供的基于多传感器的数据融合装置可执行本申请任意实施例所提供的基于多传感器的数据融合方法，具备执行方法相应的功能模块和有益效果。The multi-sensor-based data fusion device provided in the embodiment of the present application can execute the multi-sensor-based data fusion method provided in any embodiment of the present application, and has corresponding functional modules and beneficial effects for executing the method.

图5为本申请实施例提供的一种电子设备的结构示意图。图5示出了适于用来实现本申请实施方式的示例性电子设备12的框图。图5显示的电子设备12仅仅是一个示例，不应对本申请实施例的功能和使用范围带来任何限制。FIG. 5 is a schematic structural diagram of an electronic device provided by an embodiment of the present application. FIG. 5 shows a block diagram of an exemplary electronic device 12 suitable for implementing embodiments of the present application. The electronic device 12 shown in FIG. 5 is only an example, and should not limit the functions and scope of use of the embodiment of the present application.

如图5所示，电子设备12以通用计算设备的形式表现。电子设备12的组件可以包括但不限于：一个或者多个处理器或者处理单元16，***存储器28，连接不同***组件(包括***存储器28和处理单元16)的总线18。As shown in FIG. 5, electronic device 12 takes the form of a general-purpose computing device. Components of electronic device 12 may include, but are not limited to, one or more processors or processing units 16, system memory 28, bus 18 connecting various system components including system memory 28 and processing unit 16.

总线18表示几类总线结构中的一种或多种，包括存储器总线或者存储器控制器，***总线，图形加速端口，处理器或者使用多种总线结构中的任意总线结构的局域总线。举例来说，这些体系结构包括但不限于工业标准体系结构(ISA)总线，微通道体系结构(MCA)总线，增强型ISA总线、视频电子标准协会(VESA)局域总线以及***组件互连(PCI)总线。 Bus 18 represents one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processor, or a local bus using any of a variety of bus structures. Examples of these architectures include, but are not limited to, the Industry Standard Architecture (ISA) bus, the Micro Channel Architecture (MCA) bus, the Enhanced ISA bus, the Video Electronics Standards Association (VESA) local bus, and the Peripheral Component Interconnect ( PCI) bus.

电子设备12典型地包括多种计算机***可读介质。这些介质可以是任何能 够被电子设备12访问的可用介质，包括易失性和非易失性介质，可移动的和不可移动的介质。 Electronic device 12 typically includes a variety of computer system readable media. These media can be any available media that can be accessed by electronic device 12 and include both volatile and nonvolatile media, removable and non-removable media.

***存储器28可以包括易失性存储器形式的计算机***可读介质，例如随机存取存储器(RAM)30和/或高速缓存存储器32。电子设备12可以包括其它可移动/不可移动的、易失性/非易失性计算机***存储介质。仅作为举例，存储***34可以设置为读写不可移动的、非易失性磁介质(图5未显示，通常称为“硬盘驱动器”)。尽管图5中未示出，可以提供用于对可移动非易失性磁盘(例如“软盘”)读写的磁盘驱动器，以及对可移动非易失性光盘(例如CD-ROM,DVD-ROM或者其它光介质)读写的光盘驱动器。在这些情况下，每个驱动器可以通过一个或者多个数据介质接口与总线18相连。***存储器28可以包括至少一个程序产品，该程序产品具有一组(例如至少一个)程序模块，这些程序模块被配置以执行本申请多个实施例的功能。 System memory 28 may include computer system readable media in the form of volatile memory, such as random access memory (RAM) 30 and/or cache memory 32 . Electronic device 12 may include other removable/non-removable, volatile/nonvolatile computer system storage media. By way of example only, storage system 34 may be configured to read and write to non-removable, non-volatile magnetic media (not shown in FIG. 5, commonly referred to as a "hard drive"). Although not shown in FIG. 5, a disk drive for reading and writing to a removable non-volatile disk (such as a "floppy disk") may be provided, as well as a removable non-volatile disk (such as a CD-ROM, DVD-ROM or other optical media) CD-ROM drive. In these cases, each drive may be connected to bus 18 via one or more data media interfaces. System memory 28 may include at least one program product having a set (eg, at least one) of program modules configured to perform the functions of various embodiments of the present application.

具有一组(至少一个)程序模块42的程序/实用工具40，可以存储在例如***存储器28中，这样的程序模块42包括但不限于操作***、一个或者多个应用程序、其它程序模块以及程序数据，这些示例中的每一个或某种组合中可能包括网络环境的实现。程序模块42通常执行本申请所描述的实施例中的功能和/或方法。Program/utility 40 may be stored, for example, in system memory 28 as a set (at least one) of program modules 42 including, but not limited to, an operating system, one or more application programs, other program modules, and program data, each or some combination of these examples may include the realization of the network environment. The program modules 42 generally perform the functions and/or methods of the embodiments described herein.

电子设备12也可以与一个或多个外部设备14(例如键盘、指向设备、显示器24等)通信，还可与一个或者多个使得用户能与该电子设备12交互的设备通信，和/或与使得该电子设备12能与一个或多个其它计算设备进行通信的任何设备(例如网卡，调制解调器等等)通信。这种通信可以通过输入/输出(I/O)接口22进行。并且，电子设备12还可以通过网络适配器20与一个或者多个网络(例如局域网(LAN)，广域网(WAN)和/或公共网络，例如因特网)通信。如图所示，网络适配器20通过总线18与电子设备12的其它模块通信。应当明白，尽管图5中未示出，可以结合电子设备12使用其它硬件和/或软件模块，包括但不限于：微代码、设备驱动器、冗余处理单元、外部磁盘驱动阵列、RAID***、磁带驱动器以及数据备份存储***等。The electronic device 12 may also communicate with one or more external devices 14 (e.g., a keyboard, pointing device, display 24, etc.), may also communicate with one or more devices that enable a user to interact with the electronic device 12, and/or communicate with Any device (eg, network card, modem, etc.) that enables the electronic device 12 to communicate with one or more other computing devices. Such communication may occur through input/output (I/O) interface 22 . Moreover, the electronic device 12 can also communicate with one or more networks (such as a local area network (LAN), a wide area network (WAN) and/or a public network such as the Internet) through the network adapter 20 . As shown, network adapter 20 communicates with other modules of electronic device 12 via bus 18 . It should be appreciated that although not shown in FIG. 5, other hardware and/or software modules may be used in conjunction with electronic device 12, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape Drives and data backup storage systems, etc.

处理单元16通过运行存储在***存储器28中的程序，从而执行多种功能应用以及数据处理，例如实现本申请实施例所提供的基于多传感器的数据融合方法。The processing unit 16 executes various functional applications and data processing by running the programs stored in the system memory 28 , for example, implementing the multi-sensor based data fusion method provided by the embodiment of the present application.

本申请实施例还提供了一种计算机可读存储介质，其上存储有计算机程序，该程序被处理器执行时实现如本申请任意申请实施例提供的基于多传感器的数据融合方法。The embodiment of the present application also provides a computer-readable storage medium, on which a computer program is stored. When the program is executed by a processor, the multi-sensor-based data fusion method provided in any embodiment of the present application is implemented.

本申请实施例的计算机存储介质，可以采用一个或多个计算机可读的介质的任意组合。计算机可读介质可以是计算机可读信号介质或者计算机可读存储介质。计算机可读存储介质例如可以是但不限于电、磁、光、电磁、红外线、或半导体的***、装置或器件，或者任意以上的组合。计算机可读存储介质的更具体的例子(非穷举的列表)包括：具有一个或多个导线的电连接、便携式 计算机磁盘、硬盘、随机存取存储器(RAM)、只读存储器(ROM)、可擦式可编程只读存储器(EPROM或闪存)、光纤、便携式紧凑磁盘只读存储器(CD-ROM)、光存储器件、磁存储器件、或者上述的任意合适的组合。在本文件中，计算机可读存储介质可以是任何包含或存储程序的有形介质，该程序可以被指令执行***、装置或者器件使用或者与其结合使用。计算机可读存储介质可以是非暂态计算机可读存储介质。The computer storage medium in the embodiments of the present application may use any combination of one or more computer-readable media. The computer readable medium may be a computer readable signal medium or a computer readable storage medium. A computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or device, or any combination thereof. More specific examples (non-exhaustive list) of computer readable storage media include: electrical connections with one or more leads, portable computer disks, hard disks, random access memory (RAM), read only memory (ROM), Erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination of the above. In this document, a computer-readable storage medium may be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. The computer readable storage medium may be a non-transitory computer readable storage medium.

计算机可读的信号介质可以包括在基带中或者作为载波一部分传播的数据信号，其中承载了计算机可读的程序代码。这种传播的数据信号可以采用多种形式，包括但不限于电磁信号、光信号或上述的任意合适的组合。计算机可读的信号介质还可以是计算机可读存储介质以外的任何计算机可读介质，该计算机可读介质可以发送、传播或者传输用于由指令执行***、装置或者器件使用或者与其结合使用的程序。A computer readable signal medium may include a data signal carrying computer readable program code in baseband or as part of a carrier wave. Such propagated data signals may take many forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the foregoing. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device. .

计算机可读介质上包含的程序代码可以用任何适当的介质传输，包括但不限于无线、电线、光缆、RF等等，或者上述的任意合适的组合。Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

可以以一种或多种程序设计语言或其组合来编写用于执行本申请操作的计算机程序代码，所述程序设计语言包括面向对象的程序设计语言，诸如Java、Smalltalk、C++，还包括常规的过程式程序设计语言—诸如“C”语言或类似的程序设计语言。程序代码可以完全地在用户计算机上执行、部分地在用户计算机上执行、作为一个独立的软件包执行、部分在用户计算机上部分在远程计算机上执行、或者完全在远程计算机或服务器上执行。在涉及远程计算机的情形中，远程计算机可以通过任意种类的网络包括局域网(LAN)或广域网(WAN)连接到用户计算机，或者，可以连接到外部计算机(例如利用因特网服务提供商来通过因特网连接)。Computer program codes for performing the operations of the present application may be written in one or more programming languages or combinations thereof, including object-oriented programming languages such as Java, Smalltalk, C++, and conventional Procedural Programming Language - such as "C" or a similar programming language. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. Where a remote computer is involved, the remote computer may be connected to the user computer through any kind of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g. via the Internet using an Internet Service Provider). .

Claims (10)

1. 一种基于多传感器的数据融合方法，包括：A multi-sensor based data fusion method, comprising:

   获取至少两种传感器分别采集的原传感器数据；Obtaining raw sensor data collected by at least two sensors;

   针对每种传感器，根据每种传感器采集的原传感器数据进行预测，得到预测传感器数据；For each sensor, predict according to the original sensor data collected by each sensor, and obtain the predicted sensor data;

   根据所述原传感器数据和所述预测传感器数据，确定所述每种传感器的数据差值序列；According to the original sensor data and the predicted sensor data, determine the data difference sequence of each sensor;

   确定所述每种传感器的数据差值序列是否满足异常检测条件，响应于确定所述每种传感器的数据差值序列满足异常检测条件，根据所述每种传感器的数据差值序列确定所述每种传感器的融合权重；Determining whether the data difference sequence of each sensor satisfies the abnormality detection condition, in response to determining that the data difference sequence of each sensor satisfies the abnormality detection condition, determining the each sensor according to the data difference sequence of each sensor Fusion weights of various sensors;

   根据所述至少两种传感器的融合权重，对所述至少两种传感器分别采集的原传感器数据进行融合。According to the fusion weights of the at least two sensors, the original sensor data respectively collected by the at least two sensors are fused.
2. 根据权利要求1所述的方法，其中，所述确定所述每种传感器的数据差值序列是否满足异常检测条件，包括：The method according to claim 1, wherein said determining whether the data difference sequence of each sensor satisfies an abnormality detection condition comprises:

   确定所述每种传感器的数据差值序列中是否存在大于异常检测阈值的数据。It is determined whether there is data greater than an abnormality detection threshold in the data difference sequence of each sensor.
3. 根据权利要求2所述的方法，所述获取至少两种传感器分别采集的原传感器数据之后，还包括：The method according to claim 2, after said acquiring the original sensor data collected by at least two sensors respectively, further comprising:

   从所述原传感器数据中筛选出每种传感器标志位正常且状态稳定时的数据，作为所述每种传感器的目标传感器数据；Screen out the data when each sensor flag is normal and in a stable state from the original sensor data, as the target sensor data for each sensor;

   根据所述目标传感器数据，确定目标预测数据；determining target prediction data according to the target sensor data;

   根据所述目标预测数据和所述目标传感器数据，确定目标差值序列；determining a target difference sequence according to the target prediction data and the target sensor data;

   对所述目标差值序列进行分段，计算每段数据的标准差，形成目标标准差序列；Segmenting the target difference sequence, calculating the standard deviation of each segment of data to form a target standard deviation sequence;

   根据所述目标标准差序列，确定所述每种传感器的所述异常检测阈值。The abnormality detection threshold of each sensor is determined according to the target standard deviation sequence.
4. 根据权利要求3所述的方法，其中，所述根据所述目标标准差序列，确定所述每种传感器的所述异常检测阈值，包括：The method according to claim 3, wherein said determining the abnormality detection threshold of each sensor according to the target standard deviation sequence comprises:

   根据所述目标标准差序列，确定所述目标标准差序列的均值和标准差；determining the mean and standard deviation of the target standard deviation sequence according to the target standard deviation sequence;

   基于所述目标标准差序列的均值和标准差，以发生概率小于50％的原则，确定所述每种传感器的所述异常检测阈值。Based on the mean value and standard deviation of the target standard deviation sequence, the abnormality detection threshold of each sensor is determined on the principle that the probability of occurrence is less than 50%.
5. 根据权利要求1所述的方法，其中，所述根据所述每种传感器的数据差值序列确定所述每种传感器的融合权重，包括：The method according to claim 1, wherein said determining the fusion weight of each sensor according to the data difference sequence of each sensor comprises:

   对所述每种传感器的数据差值序列和目标差值序列进行一致性检测，确定一致性检测值；Perform consistency detection on the data difference sequence and the target difference sequence of each sensor, and determine the consistency detection value;

   根据所述一致性检测值，确定所述每种传感器的融合权重。According to the consistency detection value, the fusion weight of each sensor is determined.
6. 根据权利要求5所述的方法，其中，所述根据所述一致性检测值，确定所述每种传感器的融合权重，包括：The method according to claim 5, wherein said determining the fusion weight of each sensor according to said consistency detection value comprises:

   响应于确定所述一致性检测值大于一致性检测阈值，根据所述每种传感器的基础权重值和所述一致性检测值确定所述每种传感器的融合权重；In response to determining that the consistency detection value is greater than a consistency detection threshold, determining a fusion weight for each sensor based on a base weight value for each sensor and the consistency detection value;

   响应于确定所述一致性检测值小于或等于所述一致性检测阈值，根据所述 每种传感器的基础权重值确定所述每种传感器的融合权重。In response to determining that the consistency detection value is less than or equal to the consistency detection threshold, a fusion weight for each sensor is determined based on a base weight value for each sensor.
7. 根据权利要求1所述的方法，还包括：The method according to claim 1, further comprising:

   根据所述每种传感器的类型和融合属性，确定所述每种传感器的基础权重值。The basic weight value of each sensor is determined according to the type and fusion attribute of each sensor.
8. 一种基于多传感器的数据融合装置，包括：A multi-sensor based data fusion device, comprising:

   数据获取模块，设置为获取至少两种传感器分别采集的原传感器数据；The data acquisition module is configured to acquire the original sensor data collected by at least two sensors;

   数据预测模块，设置为针对每种传感器，根据每种传感器采集的原传感器数据进行预测，得到预测传感器数据；The data prediction module is configured to perform prediction according to the original sensor data collected by each sensor to obtain predicted sensor data for each sensor;

   差值确定模块，设置为根据所述原传感器数据和所述预测传感器数据，确定所述每种传感器的数据差值序列；The difference determination module is configured to determine the data difference sequence of each sensor according to the original sensor data and the predicted sensor data;

   权重确定模块，设置为确定所述传每种感器的数据差值序列是否满足异常检测条件，响应于确定所述每种传感器的数据差值序列满足异常检测条件，根据所述每种传感器的数据差值序列确定所述每种传感器的融合权重；The weight determination module is configured to determine whether the data difference sequence of each sensor satisfies the abnormality detection condition, and in response to determining that the data difference sequence of each sensor satisfies the abnormality detection condition, according to the data difference sequence of each sensor The data difference sequence determines the fusion weight of each sensor;

   数据融合模块，设置为根据所述至少两种传感器的融合权重，对所述至少两种传感器分别采集的原传感器数据进行融合。The data fusion module is configured to fuse the original sensor data respectively collected by the at least two sensors according to the fusion weights of the at least two sensors.
9. 一种电子设备，包括：An electronic device comprising:

   一个或多个处理器；one or more processors;

   存储装置，设置为存储一个或多个程序，storage means configured to store one or more programs,

   当所述一个或多个程序被所述一个或多个处理器执行，使得所述一个或多个处理器实现如权利要求1-7中任一所述的基于多传感器的数据融合方法。When the one or more programs are executed by the one or more processors, the one or more processors are made to implement the multi-sensor-based data fusion method according to any one of claims 1-7.
10. 一种计算机可读存储介质，其上存储有计算机程序，所述计算机程序被处理器执行时实现如权利要求1-7中任一所述的基于多传感器的数据融合方法。A computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the multi-sensor-based data fusion method according to any one of claims 1-7 is implemented.

Images