WO2021068378A1

WO2021068378A1 - Parking space identification and compensation method based on ultrasonic sensor parking space detection system

Info

Publication number: WO2021068378A1
Application number: PCT/CN2019/121203
Authority: WO
Inventors: 赵宇鹏; 罗作煌; 胡坤福
Original assignee: 惠州市德赛西威智能交通技术研究院有限公司
Priority date: 2019-10-12
Filing date: 2019-11-27
Publication date: 2021-04-15
Also published as: CN110853399A

Abstract

A parking space identification and compensation method based on an ultrasonic sensor parking space detection system, the method comprising: respectively acquiring, by means of an ultrasonic radar, shape types of obstacles at an upper edge and a lower edge of a target parking space, and calculating the widths of the obstacles at the upper edge and the lower edge of the target parking space, and the distance between the upper edge and the lower edge of the target parking space; respectively determining compensation values of the upper edge and the lower edge of the target parking space; and according to the distance between the upper edge and the lower edge of the target parking space, and the compensation values of the upper edge and the lower edge of the target parking space, calculating the length of the target parking space. Features of obstacles at an upper edge and a lower edge of a target parking space are acquired by means of an ultrasonic sensor, and the upper edge and the lower edge of the target parking space are compensated for in conjunction with a vehicle speed, such that the target parking space is identified accurately. According to the method, higher robustness and applicability are provided, the edge identification precision of the ultrasonic sensor can be effectively improved, and more accurate parking space decision information is provided for an automatic parking system.

Description

一种基于超声波传感器车位检测***的车位识别补偿方法Parking space recognition and compensation method based on ultrasonic sensor parking space detection system

技术领域Technical field

本发明涉及自动泊车技术领域，特别是涉及一种基于超声波传感器车位检测***的车位识别补偿方法。The invention relates to the technical field of automatic parking, in particular to a parking space recognition and compensation method based on an ultrasonic sensor parking space detection system.

背景技术Background technique

随着我国汽车保有量的逐年递增，城市公共泊车空间也变得日趋紧张，因泊车而造成的交通事故率也逐渐上升。如何提升现代城市交通中汽车的驾驶安全性及舒适性已成为各大汽车制造商及汽车电子供应商研究的重点。自动泊车***作为智能辅助驾驶***，可满足人们对于在城市窄小泊车空间实现精确泊车的需求，能够有效降低泊车事故率。车位识别作为自动泊车***中重要的感知环节，其对泊车精度有着至关重要的影响。如何提高车位识别精度也是提升自动泊车***性能的关键。As the number of cars in my country has increased year by year, urban public parking spaces have become increasingly tense, and the rate of traffic accidents caused by parking has gradually increased. How to improve the driving safety and comfort of automobiles in modern urban traffic has become the research focus of major automobile manufacturers and automotive electronics suppliers. As an intelligent driving assistance system, the automatic parking system can meet people's needs for accurate parking in small urban parking spaces, and can effectively reduce the rate of parking accidents. As an important perception link in the automatic parking system, parking space recognition has a vital impact on parking accuracy. How to improve the accuracy of parking space recognition is also the key to improving the performance of the automatic parking system.

目前市场现有的车位识别技术主要有三种：利用AVM(360全景环视***)感知当前车辆周围环境信息以及与目标车位间的相对位置信息；利用视觉传感器以及超声波传感器信息融合的方式，在目标车位库口进行初始寻库偏差矫正的方法；利用超声波传感器感知车辆与障碍物之间的距离信息，计算判断得出可泊车位信息。At present, there are mainly three kinds of parking space recognition technologies available in the market: using AVM (360 panoramic surround view system) to perceive the current surrounding environment information of the vehicle and the relative position information with the target parking space; using visual sensors and ultrasonic sensor information fusion method, in the target parking space The method of initial search deviation correction at the warehouse entrance; the use of ultrasonic sensors to perceive the distance information between the vehicle and the obstacle, and the calculation and judgment to obtain the parking space information.

但这三种方式均有不足之处，对于AVM***它仅能获取目标车位的占用情况，而无法感知车位内的障碍物(锥形交通路标、车位锁等)信息；而泊车初始寻库偏差矫正法则很难适应实际中复杂的泊车环境，因而无法保证其鲁棒性，且标定工作量巨大；而依靠超声波传感器的车位识别技术，由于超声波传感器单发单收的收发机制，在车位识别过程中目标车位的上、下边缘识别精度主要受限于车速以及障碍物形状的影响而存在较大的识别误差；虽然该技术易落地、成本低、经济效益好，但如何克服超声波传感器测距特性造成的车位边缘识别误差大的问题是其核心技术难题。But these three methods have shortcomings. For the AVM system, it can only obtain the occupancy status of the target parking space, and cannot sense the information of the obstacles (cone traffic signs, parking locks, etc.) in the parking space; and the initial parking search The deviation correction method is difficult to adapt to the actual complex parking environment, so its robustness cannot be guaranteed, and the calibration workload is huge; while relying on the parking space recognition technology of the ultrasonic sensor, due to the ultrasonic sensor's single-transmitting and single-receiving transmission and reception mechanism, in the parking space During the recognition process, the recognition accuracy of the upper and lower edges of the target parking space is mainly limited by the influence of the vehicle speed and the shape of the obstacle, and there is a large recognition error; although this technology is easy to land, low cost, and good economic benefits, how to overcome the ultrasonic sensor measurement The problem of large margin recognition error caused by distance characteristics is its core technical problem.

针对以上车位识别技术所面临的问题，本文提出了一种基于超声波传感器自动泊车***的车位识别补偿计算方法，以解决超声波传感器在车位识别过程中对于车位边缘识别误差大、精度低的问题，有效降低了标定成本，提升了车位识别***的鲁棒性。In view of the problems faced by the above parking space recognition technology, this paper proposes a parking space recognition compensation calculation method based on the ultrasonic sensor automatic parking system to solve the problem of large error and low accuracy of the edge recognition of the parking space by the ultrasonic sensor in the parking space recognition process. This effectively reduces the cost of calibration and improves the robustness of the parking space recognition system.

发明内容Summary of the invention

本发明为克服上述现有技术所述的不足，提供一种基于超声波传感器车位检测***的车位识别补偿方法。In order to overcome the above-mentioned shortcomings of the prior art, the present invention provides a parking space recognition and compensation method based on an ultrasonic sensor parking space detection system.

为解决上述技术问题，本发明的技术方案如下：In order to solve the above technical problems, the technical scheme of the present invention is as follows:

一种基于超声波传感器车位检测***的车位识别补偿方法，包括如下步骤：A parking space recognition and compensation method based on an ultrasonic sensor parking space detection system includes the following steps:

识别当前车速，并获取多帧超声波雷达测距数据；Identify the current vehicle speed and obtain multiple frames of ultrasonic radar ranging data;

根据获取的超声波雷达测距数据分别计算目标车位上边缘和目标车位下边缘的障碍物宽度、目标车位上边缘与目标车位下边缘之间的距离；Calculate the obstacle width of the upper edge of the target parking space and the lower edge of the target parking space, and the distance between the upper edge of the target parking space and the lower edge of the target parking space according to the acquired ultrasonic radar ranging data;

根据获取的超声波雷达测距数据分别获取目标车位上边缘和目标车位下边缘的障碍物形状类型；Obtain the obstacle shape types at the upper edge of the target parking space and the lower edge of the target parking space according to the acquired ultrasonic radar ranging data;

根据车速、目标车位上边缘的障碍物宽度和目标车位上边缘的障碍物形状类型确定目标车位上边缘的补偿值；Determine the compensation value of the upper edge of the target parking space according to the vehicle speed, the width of the obstacle on the upper edge of the target parking space and the obstacle shape type on the upper edge of the target parking space;

根据车速、目标车位下边缘的障碍物宽度和目标车位下边缘的障碍物形状类型确定目标车位下边缘的补偿值；Determine the compensation value of the lower edge of the target parking space according to the vehicle speed, the width of the obstacle at the lower edge of the target parking space and the obstacle shape type at the lower edge of the target parking space;

根据目标车位上边缘与目标车位下边缘之间的距离、目标车位上边缘的补偿值和目标车位下边缘的补偿值计算目标车位长度。The length of the target parking space is calculated according to the distance between the upper edge of the target parking space and the lower edge of the target parking space, the compensation value of the upper edge of the target parking space and the compensation value of the lower edge of the target parking space.

进一步的，作为优选技术方案，障碍物形状类型的获取具体包括：Further, as a preferred technical solution, the acquisition of the obstacle shape type specifically includes:

获取多帧超声波雷达测距数据，并对超声波雷达测距数据进行滤波处理；Acquire multiple frames of ultrasonic radar ranging data, and filter the ultrasonic radar ranging data;

对处理后的超声波雷达测距数据进行缓存并计算每帧超声波雷达测距数据权重值；Cache the processed ultrasonic radar ranging data and calculate the weight value of each frame of ultrasonic radar ranging data;

根据每帧超声波雷达测距数据权重值及连续获取的三帧超声波雷达测距数据之间的变化情况对缓存的超声波雷达测距数据进行处理得到有效超声波雷达测距数据；According to the weight value of each frame of ultrasonic radar ranging data and the changes between the three consecutive frames of ultrasonic radar ranging data, the buffered ultrasonic radar ranging data is processed to obtain effective ultrasonic radar ranging data;

提取有效超声波雷达测距数据所构建的障碍物轮廓特征，从而得到障碍物形状类型。Extract the obstacle contour features constructed by the effective ultrasonic radar ranging data to obtain the obstacle shape type.

进一步的，作为优选技术方案，获取有效超声波雷达测距数据具体包括：Further, as a preferred technical solution, obtaining effective ultrasonic radar ranging data specifically includes:

分别判断每帧超声波雷达测距数据权重值是否大于第一预设阈值，若是，执行下一步，否则，从缓存的超声波雷达测距数据中剔除该帧超声波雷达测距数据；Respectively determine whether the weight value of each frame of ultrasonic radar ranging data is greater than the first preset threshold, if yes, proceed to the next step, otherwise, remove the frame of ultrasonic radar ranging data from the cached ultrasonic radar ranging data;

判断连续获取的三帧超声波雷达测距数据是否超出超声波雷达测距范围，若是，从缓存的超声波雷达测距数据中剔除该帧超声波雷达测距数据，否则，执行下一步；Determine whether the three consecutive frames of ultrasonic radar ranging data are beyond the ultrasonic radar ranging range, if so, remove the frame of ultrasonic radar ranging data from the cached ultrasonic radar ranging data, otherwise, proceed to the next step;

判断获取的当前帧超声波雷达测距数据是否大于上一帧超声波雷达测距数据，同时判断获取的当前帧超声波雷达测距数据是否大于下一帧超声波雷达测距数据，若是，从缓存的超声波雷达测距数据中剔除该帧超声波雷达测距数据，否则，执行下一步；Determine whether the acquired current frame of ultrasonic radar ranging data is greater than the previous frame of ultrasonic radar ranging data, and at the same time determine whether the acquired current frame of ultrasonic radar ranging data is greater than the next frame of ultrasonic radar ranging data, if so, from the cached ultrasonic radar Eliminate this frame of ultrasonic radar ranging data from the ranging data, otherwise, proceed to the next step;

判断获取的当前帧超声波雷达测距数据是否小于上一帧超声波雷达测距数据，同时判断获取的当前帧超声波雷达测距数据是否小于下一帧超声波雷达测距数据，若是，从缓存的超声波雷达测距数据中剔除该帧超声波雷达测距数据，否则，执行下一步；Determine whether the acquired current frame of ultrasonic radar ranging data is less than the previous frame of ultrasonic radar ranging data, and at the same time determine whether the acquired current frame of ultrasonic radar ranging data is less than the next frame of ultrasonic radar ranging data, if so, from the cached ultrasonic radar Eliminate this frame of ultrasonic radar ranging data from the ranging data, otherwise, proceed to the next step;

保留当前帧超声波雷达测距数据，从而得到有效超声波雷达测距数据。The ultrasonic radar ranging data of the current frame is retained to obtain effective ultrasonic radar ranging data.

进一步的，作为优选技术方案，在计算每帧超声波雷达测距数据的权重值之前，需先判断所缓存的超声波雷达测距数据的数量是否大于第二预设阈值，若是，计算每帧超声波雷达测距数据的权重值，否则，重新获取超声波雷达测距数据。Further, as a preferred technical solution, before calculating the weight value of each frame of ultrasonic radar ranging data, it is necessary to determine whether the number of buffered ultrasonic radar ranging data is greater than the second preset threshold, if so, calculate each frame of ultrasonic radar The weight value of the ranging data, otherwise, reacquire the ultrasonic radar ranging data.

进一步的，作为优选技术方案，所述第一预设阈值的设置范围为0.3-0.7，所述第二预设阈值的设置范围为3-7。Further, as a preferred technical solution, the setting range of the first preset threshold is 0.3-0.7, and the setting range of the second preset threshold is 3-7.

进一步的，作为优选技术方案，每帧超声波雷达测距数据权重值的计算具体包括：Further, as a preferred technical solution, the calculation of the weight value of each frame of ultrasonic radar ranging data specifically includes:

缓存的超声波雷达测距数据中最大权重值和每帧超声波雷达测距数据的探测次数与测距概率因数的乘积之比。The ratio of the maximum weight value in the buffered ultrasonic radar ranging data to the product of the number of times of detection of each frame of ultrasonic radar ranging data and the ranging probability factor.

进一步的，作为优选技术方案，障碍物形状类型包括：方形、圆形和未知型。Further, as a preferred technical solution, the types of obstacle shapes include: square, round, and unknown.

进一步的，作为优选技术方案，障碍物宽度的获取具体包括：Further, as a preferred technical solution, obtaining the width of the obstacle specifically includes:

识别障碍物起始边缘并记录当前时刻为第一时刻；Identify the starting edge of the obstacle and record the current moment as the first moment;

识别障碍物终止边缘并记录当前时刻为第二时刻；Identify the end edge of the obstacle and record the current moment as the second moment;

计算车辆在第一时刻到第二时刻之间的行驶距离为障碍物宽度。Calculate the distance of the vehicle from the first moment to the second moment as the width of the obstacle.

进一步的，作为优选技术方案，还包括：Further, as a preferred technical solution, it also includes:

识别障碍物起始边缘，判断当前帧所获取的超声波雷达测距数据与上一帧所获取的超声波雷达测距数据之间的变化值是否大于第三预设阈值，若是，记录当前时刻为第一时刻，否则，重新识别障碍物起始边缘；Identify the starting edge of the obstacle, and determine whether the change value between the ultrasonic radar ranging data acquired in the current frame and the ultrasonic radar ranging data acquired in the previous frame is greater than the third preset threshold. If so, record the current moment as the first For a moment, otherwise, re-identify the starting edge of the obstacle;

识别障碍物终止边缘，判断当前帧所获取的超声波雷达测距数据与上一帧所获取的超声波雷达测距数据之间的变化值是否大于第三预设阈值，若是，记录当前时刻为第二时刻，否则，重新识别障碍物终止边缘。Identify the end edge of the obstacle, determine whether the change value between the ultrasonic radar ranging data obtained in the current frame and the ultrasonic radar ranging data obtained in the previous frame is greater than the third preset threshold, if so, record the current moment as the second At any time, otherwise, re-identify the obstacle termination edge.

进一步的，作为优选技术方案，所述第三预设阈值根据车辆宽度设置。Further, as a preferred technical solution, the third preset threshold is set according to the width of the vehicle.

进一步的，作为优选技术方案，确定目标车位上边缘的补偿值和目标车位下边缘的补偿值具体包括：Further, as a preferred technical solution, determining the compensation value of the upper edge of the target parking space and the compensation value of the lower edge of the target parking space specifically includes:

根据目标车位上边缘的障碍物宽度确定目标车位上边缘补偿系数，根据车速和目标车位上边缘的障碍物形状类型确定目标车位上边缘补偿参数，目标车位上边缘的补偿值为目标车位上边缘补偿系数与目标车位上边缘补偿参数的乘积；Determine the upper edge compensation coefficient of the target parking space according to the obstacle width of the upper edge of the target parking space, and determine the upper edge compensation parameters of the target parking space according to the vehicle speed and the obstacle shape type of the upper edge of the target parking space. The compensation value of the upper edge of the target parking space is the upper edge compensation of the target parking space. The product of the coefficient and the edge compensation parameter of the target parking space;

根据目标车位下边缘的障碍物宽度确定目标车位下边缘补偿系数，根据车速和目标车位下边缘的障碍物形状类型确定目标车位下边缘补偿参数，目标车位下边缘的补偿值为目标车位下边缘补偿系数与目标车位下边缘补偿参数的乘积。Determine the lower edge compensation coefficient of the target parking space according to the obstacle width of the lower edge of the target parking space, and determine the lower edge compensation parameters of the target parking space according to the vehicle speed and the obstacle shape type of the lower edge of the target parking space. The compensation value of the lower edge of the target parking space is the bottom edge compensation of the target parking space. The product of the coefficient and the bottom edge compensation parameter of the target parking space.

进一步的，作为优选技术方案，目标车位上边缘补偿系数和目标车位下边缘补偿系数的获取具体包括：Further, as a preferred technical solution, the acquisition of the upper edge compensation coefficient of the target parking space and the lower edge compensation coefficient of the target parking space specifically includes:

补偿系数的设置随障碍物宽度的增大而减小。The setting of the compensation coefficient decreases as the width of the obstacle increases.

进一步的，作为优选技术方案，目标车位上边缘补偿参数和目标车位下边缘补偿参数的获取具体包括：Further, as a preferred technical solution, the acquisition of the upper edge compensation parameter of the target parking space and the lower edge compensation parameter of the target parking space specifically includes:

补偿参数的设置随车速的增大以及障碍物形状类型所对应的边缘曲率的增大而增大。The setting of the compensation parameter increases with the increase of the vehicle speed and the increase of the edge curvature corresponding to the obstacle shape type.

进一步的，作为优选技术方案，目标车位长度通过以下方式计算：Further, as a preferred technical solution, the length of the target parking space is calculated in the following way:

目标车位长度为目标车位上边缘与目标车位下边缘之间的距离、目标车位上边缘的补偿值和目标车位下边缘的补偿值求和。The length of the target parking space is the sum of the distance between the upper edge of the target parking space and the lower edge of the target parking space, the compensation value of the upper edge of the target parking space and the compensation value of the lower edge of the target parking space.

与现有技术相比，本发明技术方案的有益效果是：Compared with the prior art, the technical solution of the present invention has the following beneficial effects:

本发明通过超声波传感器获取目标车位上边缘和目标车位下边缘的障碍物形状类型以及宽度，同时结合车速对超声波传感器测距计算的目标车位的上边缘和下边缘进行补偿计算，从而识别出准确的目标车位，该识别方法具有更强的鲁棒性、适用性，可有效提高超声波传感器的边缘识别精度，为自动泊车***提供更加精确地车位决策信息。The invention obtains the obstacle shape type and width of the upper edge of the target parking space and the lower edge of the target parking space through the ultrasonic sensor, and simultaneously performs compensation calculation on the upper and lower edges of the target parking space calculated by the ultrasonic sensor distance measurement in combination with the vehicle speed, thereby identifying accurate The target parking space, the recognition method has stronger robustness and applicability, can effectively improve the edge recognition accuracy of the ultrasonic sensor, and provide more accurate parking space decision information for the automatic parking system.

附图说明Description of the drawings

图1为本发明方法步骤流程图。Figure 1 is a flow chart of the steps of the method of the present invention.

图2为本发明障碍物宽度的计算方法步骤流程图。Figure 2 is a flow chart of the steps of the method for calculating the width of the obstacle of the present invention.

图2为本发明障碍物形状类型的计算方法步骤流程图。Figure 2 is a flow chart of the steps of the method for calculating the shape of the obstacle of the present invention.

附图仅用于示例性说明，不能理解为对本专利的限制；为了更好说明本实施例，附图某些部件会有省略、放大或缩小，并不代表实际产品的尺寸；对于本领域技术人员来说，附图中某些公知结构及其说明可能省略是可以理解的；相同或相似的标号对应相同或相似的部件；附图中描述位置关系的用语仅用于示例性说明，不能理解为对本专利的限制。The attached drawings are only for illustrative purposes and cannot be understood as a limitation of this patent; in order to better illustrate this embodiment, some parts of the attached drawings may be omitted, enlarged or reduced, and do not represent the size of the actual product; For the personnel, it is understandable that some well-known structures and their descriptions in the drawings may be omitted; the same or similar reference numerals correspond to the same or similar parts; the terms describing the positional relationship in the drawings are only for exemplary description and cannot be understood. It is a limitation of this patent.

具体实施方式Detailed ways

下面结合附图对本发明的较佳实施例进行详细阐述，以使本发明的优点和特征更易被本领域技术人员理解，从而对本发明的保护范围作出更为清楚的界定。The preferred embodiments of the present invention will be described in detail below with reference to the accompanying drawings, so that the advantages and features of the present invention can be more easily understood by those skilled in the art, so as to make a clearer definition of the protection scope of the present invention.

实施例1Example 1

一种基于超声波传感器车位检测***的车位识别补偿方法，如图1所示：包括如下步骤：A parking space recognition and compensation method based on an ultrasonic sensor parking space detection system, as shown in Figure 1: It includes the following steps:

S10.识别当前车速，并获取多帧超声波雷达测距数据。S10. Identify the current vehicle speed and obtain multiple frames of ultrasonic radar ranging data.

在本步骤中，将车速划分为6阶段，分别为0～5km/h，5～10km/h，10～15km/h，15～20km/h，20～25km/h，25～30km/h。In this step, the vehicle speed is divided into 6 stages, which are 0-5km/h, 5-10km/h, 10-15km/h, 15-20km/h, 20-25km/h, and 25-30km/h.

本步骤具体为：识别当前车速，并获取当前车速下的多帧超声波雷达测距数据。This step is specifically: identifying the current vehicle speed, and acquiring multiple frames of ultrasonic radar ranging data at the current vehicle speed.

S20.根据获取的超声波雷达测距数据计算目标车位上边缘的障碍物宽度以及目标车位上边缘的补偿系数。S20. Calculate the obstacle width on the upper edge of the target parking space and the compensation coefficient of the upper edge of the target parking space according to the acquired ultrasonic radar ranging data.

本步骤中，目标车位上边缘的障碍物宽度的计算如图2所示，具体包括：In this step, the calculation of the obstacle width at the upper edge of the target parking space is shown in Figure 2, which specifically includes:

S201.识别目标车位上边缘的障碍物起始边缘并记录当前时刻为第一时刻。S201. Identify the starting edge of the obstacle at the upper edge of the target parking space and record the current moment as the first moment.

本步骤具体为：识别目标车位上边缘的障碍物起始边缘，判断当前帧所获取的超声波雷达测距数据与上一帧所获取的超声波雷达测距数据之间的变化值是否大于第三预设阈值，即，判断当前帧所获取的超声波雷达测距数据与上一帧所获取的超声波雷达测距数据之间的差值是否大于第三预设阈值，若是，记录当前时刻为第一时刻，否则，重新识别目标车位上边缘的障碍物起始边缘。This step is specifically: identifying the starting edge of the obstacle at the upper edge of the target parking space, and judging whether the change value between the ultrasonic radar ranging data acquired in the current frame and the ultrasonic radar ranging data acquired in the previous frame is greater than the third prediction. Set the threshold, that is, determine whether the difference between the ultrasonic radar ranging data acquired in the current frame and the ultrasonic radar ranging data acquired in the previous frame is greater than the third preset threshold, if so, record the current moment as the first moment Otherwise, re-identify the starting edge of the obstacle on the upper edge of the target parking space.

S202.识别目标车位上边缘的障碍物终止边缘并记录当前时刻为第二时刻。S202. Identify the obstacle termination edge on the upper edge of the target parking space and record the current moment as the second moment.

本步骤具体为：识别目标车位上边缘的障碍物终止边缘，判断当前帧所获取的超声波雷达测距数据与上一帧所获取的超声波雷达测距数据之间的变化值是否大于第三预设阈值，即，判断上一帧所获取的超声波雷达测距数据与当前帧所获取的超声波雷达测距数据之间的差值是否大于第三预设阈值，若是，记录当前时刻为第二时刻，否则，重新识别目标车位上边缘的障碍物终止边缘。This step is specifically: identifying the obstacle termination edge on the upper edge of the target parking space, and judging whether the change value between the ultrasonic radar ranging data obtained in the current frame and the ultrasonic radar ranging data obtained in the previous frame is greater than the third preset Threshold, that is, to determine whether the difference between the ultrasonic radar ranging data acquired in the previous frame and the ultrasonic radar ranging data acquired in the current frame is greater than the third preset threshold, if so, record the current moment as the second moment, Otherwise, re-identify the obstacle termination edge on the upper edge of the target parking space.

其中，第三预设阈值根据车辆宽度设置，其设置范围为车辆宽度与一固定值之和，在本实施例中，该固定值设置范围为0.6-1.0，优选的，固定值设置为0.8。The third preset threshold is set according to the width of the vehicle, and its setting range is the sum of the width of the vehicle and a fixed value. In this embodiment, the fixed value setting range is 0.6-1.0, and preferably, the fixed value is set to 0.8.

S203.计算车辆在第一时刻到第二时刻之间的行驶距离为目标车位上边缘的障碍物宽度。S203. Calculate the travel distance of the vehicle from the first moment to the second moment as the obstacle width at the upper edge of the target parking space.

其中，目标车位上边缘的障碍物宽度为车速与第二时刻与第一时刻时差乘积。Wherein, the width of the obstacle on the upper edge of the target parking space is the product of the vehicle speed and the time difference between the second time and the first time.

举例说明，目标车位上边缘的障碍物宽度通过以下公式计算：For example, the width of the obstacle on the upper edge of the target parking space is calculated by the following formula:

W1＝V*(T2-T1)W1=V*(T2-T1)

其中，W1表示目标车位上边缘的障碍物宽度，V表示车速，T1表示第一时刻，T2表示第二时刻。Among them, W1 represents the obstacle width at the upper edge of the target parking space, V represents the vehicle speed, T1 represents the first moment, and T2 represents the second moment.

S204.根据目标车位上边缘的障碍物宽度确定目标车位上边缘的补偿系数。S204. Determine the compensation coefficient of the upper edge of the target parking space according to the width of the obstacle on the upper edge of the target parking space.

在目标车位识别过程中，由于目标车位障碍物的宽度越窄小，越易引起目标车位边缘识别误差较大，因此，在本步骤中，该补偿系数的设置与障碍物宽度成反比，即补偿系数的设置随障碍物宽度的增大而减小。In the process of identifying the target parking space, the narrower the width of the obstacle in the target parking space is, the more likely it is to cause a larger error in the edge recognition of the target parking space. Therefore, in this step, the compensation coefficient is set inversely proportional to the width of the obstacle, that is, compensation The setting of the coefficient decreases as the width of the obstacle increases.

S30.根据获取的超声波雷达测距数据计算目标车位上边缘的障碍物形状类型以及目标车位上边缘的补偿参数。S30. Calculate the obstacle shape type on the upper edge of the target parking space and the compensation parameter of the upper edge of the target parking space according to the acquired ultrasonic radar ranging data.

本步骤中，目标车位上边缘的障碍物形状类型的获取如图3所示，具体包括：In this step, the acquisition of the obstacle shape type at the upper edge of the target parking space is shown in Figure 3, which specifically includes:

S301.获取多帧超声波雷达测距数据，并对超声波雷达测距数据进行滤波处理。S301. Acquire multiple frames of ultrasonic radar ranging data, and filter the ultrasonic radar ranging data.

此步骤中，多帧超声波雷达测距数据为第一时刻到第二时刻之间所获取的。In this step, the multiple frames of ultrasonic radar ranging data are acquired between the first time and the second time.

S302.对处理后的超声波雷达测距数据进行缓存并计算每帧超声波雷达测距数据权重值。S302. Buffer the processed ultrasonic radar ranging data and calculate the weight value of each frame of ultrasonic radar ranging data.

本步骤中，在计算每帧超声波雷达测距数据权重值之前，需先判断所缓存的超声波雷达测距数据的数量是否大于第二预设阈值，若是，计算每帧超声波雷达测距数据的权重值，否则，重新获取超声波雷达测距数据。其中，第二预设阈值的设置范围为3-7，优选的，第二预设阈值最小设置为3。In this step, before calculating the weight value of each frame of ultrasonic radar ranging data, it is necessary to determine whether the number of buffered ultrasonic radar ranging data is greater than the second preset threshold, if so, calculate the weight of each frame of ultrasonic radar ranging data Value, otherwise, reacquire the ultrasonic radar ranging data. Wherein, the setting range of the second preset threshold is 3-7, and preferably, the minimum setting of the second preset threshold is 3.

本步骤中，每帧超声波雷达测距数据权重值的计算具体包括：In this step, the calculation of the weight value of each frame of ultrasonic radar ranging data specifically includes:

缓存的超声波雷达测距数据中最大权重值和每帧超声波雷达测距数据的探测次数与测距概率因数的乘积之比。The ratio of the maximum weight value in the buffered ultrasonic radar ranging data to the product of the number of detections per frame of ultrasonic radar ranging data and the ranging probability factor.

举例说明，每帧超声波雷达测距数据权重值通过以下公式计算：For example, the weight value of each frame of ultrasonic radar ranging data is calculated by the following formula:

其中，WeightCoe表示每帧超声波雷达测距数据权重值，MaxWeightCoefficient表示超声波雷达测距数据中最大权重值，ObjDetProbabilityPercentage表示测距概率因数，DtdTimes表示每帧超声波雷达测距数据的探测次数。Among them, WeightCoe represents the weight value of each frame of ultrasonic radar ranging data, MaxWeightCoefficient represents the maximum weight value in the ultrasonic radar ranging data, ObjDetProbabilityPercentage represents the ranging probability factor, and DtdTimes represents the number of detections per frame of ultrasonic radar ranging data.

S303.根据每帧超声波雷达测距数据权重值及连续获取的三帧超声波雷达测距数据之间的变化情况对缓存的超声波雷达测距数据进行处理得到有效超声波雷达测距数。S303. Process the buffered ultrasonic radar ranging data according to the weight value of each frame of ultrasonic radar ranging data and the changes between the three consecutive frames of ultrasonic radar ranging data to obtain the effective ultrasonic radar ranging number.

本步骤具体为：This step is specifically as follows:

S3031.分别判断每帧超声波雷达测距数据权重值是否大于第一预设阈值，若是，执行S3032，否则，执行S3035；S3031. Determine whether the weight value of each frame of ultrasonic radar ranging data is greater than the first preset threshold, if so, execute S3032, otherwise, execute S3035;

S3032.判断连续获取的三帧超声波雷达测距数据是否超出超声波雷达测距范围，若是，执行S3035，否则，执行S3036；S3032. Determine whether the three consecutive frames of ultrasonic radar ranging data are beyond the ultrasonic radar ranging range, if yes, execute S3035, otherwise, execute S3036;

S3033.判断获取的当前帧超声波雷达测距数据是否大于上一帧超声波雷达测距数据，同时判断获取的当前帧超声波雷达测距数据是否大于下一帧超声波雷达测距数据，若是，执行S3035，否则，执行S3036；S3033. Determine whether the acquired current frame of ultrasonic radar ranging data is greater than the previous frame of ultrasonic radar ranging data, and determine whether the acquired current frame of ultrasonic radar ranging data is greater than the next frame of ultrasonic radar ranging data, if yes, execute S3035, Otherwise, execute S3036;

S3034.判断获取的当前帧超声波雷达测距数据是否小于上一帧超声波雷达测距数据，同时判断获取的当前帧超声波雷达测距数据是否小于下一帧超声波雷达测距数据；若是，执行S3035，否则，执行S3036；S3034. Determine whether the acquired current frame of ultrasonic radar ranging data is less than the previous frame of ultrasonic radar ranging data, and determine whether the acquired current frame of ultrasonic radar ranging data is less than the next frame of ultrasonic radar ranging data; if so, execute S3035, Otherwise, execute S3036;

S3035.从缓存的超声波雷达测距数据中剔除该帧超声波雷达测距数据；S3035. Remove the frame of ultrasonic radar ranging data from the buffered ultrasonic radar ranging data;

S3036.保留当前帧超声波雷达测距数据，从而得到有效超声波雷达测距数据。S3036. Keep the ultrasonic radar ranging data of the current frame, so as to obtain effective ultrasonic radar ranging data.

在本步骤中，第一预设阈值的设置范围为0.3-0.7，优选的第一预设阈值设置为0.5；In this step, the setting range of the first preset threshold is 0.3-0.7, and the preferred first preset threshold is set to 0.5;

同时，本步骤中还包括，当当前帧超声波雷达测距数据大于等于上一帧超声波雷达测距数据，同时当前帧超声波雷达测距数据小于等于下一帧超声波雷达测距数据，或者，当前帧超声波雷达测距数据小于等于上一帧超声波雷达测距数据，同时当前帧超声波雷达测距数据大于等于下一帧超声波雷达测距数据时，保留当前帧超声波雷达测距数据，从而得到有效超声波雷达测距数据。At the same time, this step also includes, when the current frame of ultrasonic radar ranging data is greater than or equal to the previous frame of ultrasonic radar ranging data, and the current frame of ultrasonic radar ranging data is less than or equal to the next frame of ultrasonic radar ranging data, or the current frame When the ultrasonic radar ranging data is less than or equal to the previous frame of ultrasonic radar ranging data, and the current frame of ultrasonic radar ranging data is greater than or equal to the next frame of ultrasonic radar ranging data, the current frame of ultrasonic radar ranging data is retained to obtain an effective ultrasonic radar Ranging data.

在本步骤中，当超出超声波雷达测距范围时，所获取的超声波雷达测距数据为超声波雷达测距默认值。In this step, when the ultrasonic radar ranging range is exceeded, the acquired ultrasonic radar ranging data is the ultrasonic radar ranging default value.

S304.提取有效超声波雷达测距数据所构建的障碍物轮廓特征，从而得到目标车位上边缘的障碍物形状类型，同时输出该障碍物形状类型。S304. Extracting the obstacle contour features constructed by the effective ultrasonic radar ranging data, thereby obtaining the obstacle shape type at the upper edge of the target parking space, and outputting the obstacle shape type at the same time.

本步骤中，障碍物形状类型包括：方形、圆形和不规则形。In this step, the types of obstacle shapes include: square, round, and irregular.

S305.根据目标车位上边缘的障碍物的形状类型确定目标车位上边缘的补偿参数。S305. Determine the compensation parameter of the upper edge of the target parking space according to the shape type of the obstacle on the upper edge of the target parking space.

在目标车位识别过程中，由于车速越快，通常识别出的目标车位长度越小，其边缘曲率较大的目标车位障碍物往往会引起较大的目标车位边缘识别误差，因此，在本步骤中，该补偿参数的设置与车速和障碍物形状类型所对应的边缘曲率呈正比，即补偿参数的设置随车速的增大以及障碍物形状类型所对应的边缘曲率的增大而增大。In the process of target parking space recognition, because the faster the vehicle speed, the shorter the length of the target parking space is usually recognized. The obstacles of the target parking space with larger edge curvature will often cause larger target parking space edge recognition errors. Therefore, in this step The setting of the compensation parameter is proportional to the edge curvature corresponding to the vehicle speed and the obstacle shape type, that is, the setting of the compensation parameter increases with the increase of the vehicle speed and the edge curvature corresponding to the obstacle shape type.

S40.确定目标车位上边缘的补偿值。S40. Determine the compensation value of the upper edge of the target parking space.

本步骤中，目标车位上边缘的补偿值为目标车位上边缘补偿系数与目标车位上边缘补偿参数的乘积。In this step, the compensation value of the upper edge of the target parking space is the product of the upper edge compensation coefficient of the target parking space and the upper edge compensation parameter of the target parking space.

S50.根据获取的超声波雷达测距数据计算目标车位下边缘的障碍物宽度以及目标车位下边缘的补偿系数。S50. Calculate the width of the obstacle at the lower edge of the target parking space and the compensation coefficient of the lower edge of the target parking space according to the acquired ultrasonic radar ranging data.

本步骤中，目标车位下边缘的障碍物宽度的计算与目标车位上边缘的障碍物宽度的计算方式相同，如图2所示，具体包括：In this step, the calculation of the obstacle width at the lower edge of the target parking space is the same as the calculation of the obstacle width at the upper edge of the target parking space, as shown in Figure 2, which specifically includes:

S501.识别目标车位下边缘的障碍物起始边缘并记录当前时刻为第三时刻。S501. Identify the starting edge of the obstacle at the lower edge of the target parking space and record the current moment as the third moment.

本步骤具体为：识别目标车位下边缘的障碍物起始边缘，判断当前帧所获取的超声波雷达测距数据与上一帧所获取的超声波雷达测距数据之间的变化值是否大于第三预设阈值，即，判断上一帧所获取的超声波雷达测距数据与当前帧所获取的超声波雷达测距数据之间的差值是否大于第三预设阈值，若是，记录当前时刻为第三时刻，否则，重新识别目标车位下边缘的障碍物起始边缘。This step is specifically: identifying the starting edge of the obstacle at the lower edge of the target parking space, and judging whether the change value between the ultrasonic radar ranging data acquired in the current frame and the ultrasonic radar ranging data acquired in the previous frame is greater than the third prediction. Set a threshold, that is, determine whether the difference between the ultrasonic radar ranging data acquired in the previous frame and the ultrasonic radar ranging data acquired in the current frame is greater than the third preset threshold, if so, record the current moment as the third moment Otherwise, re-identify the starting edge of the obstacle at the lower edge of the target parking space.

S502.识别目标车位下边缘的障碍物终止边缘并记录当前时刻为第四时刻。S502. Identify the obstacle termination edge at the lower edge of the target parking space and record the current moment as the fourth moment.

本步骤具体为：识别目标车位下边缘的障碍物终止边缘，判断当前帧所获取的超声波雷达测距数据与上一帧所获取的超声波雷达测距数据之间的变化值是否大于第三预设阈值，即，判断当前帧所获取的超声波雷达测距数据与上一帧所获取的超声波雷达测距数据之间的差值是否大于第三预设阈值，若是，记录当前时刻为第四时刻，否则，重新识别目标车位下边缘的障碍物终止边缘。This step is specifically: identifying the obstacle termination edge at the lower edge of the target parking space, and judging whether the change value between the ultrasonic radar ranging data acquired in the current frame and the ultrasonic radar ranging data acquired in the previous frame is greater than the third preset Threshold, that is, to determine whether the difference between the ultrasonic radar ranging data acquired in the current frame and the ultrasonic radar ranging data acquired in the previous frame is greater than the third preset threshold, if so, record the current moment as the fourth moment, Otherwise, re-identify the obstacle termination edge at the lower edge of the target parking space.

其中，第三预设阈值根据车辆宽度设置，其设置范围为车辆宽度与一固定值之和，在本实施例中，该固定值设置为0.8。The third preset threshold is set according to the width of the vehicle, and its setting range is the sum of the width of the vehicle and a fixed value. In this embodiment, the fixed value is set to 0.8.

S503.计算车辆在第三时刻到第四时刻之间的行驶距离为目标车位下边缘的障碍物宽度。S503. Calculate the travel distance of the vehicle between the third moment and the fourth moment as the obstacle width at the lower edge of the target parking space.

其中，目标车位下边缘的障碍物宽度为车速与第四时刻与第三时刻时差乘积。Among them, the width of the obstacle at the lower edge of the target parking space is the product of the vehicle speed and the time difference between the fourth time and the third time.

举例说明，目标车位下边缘的障碍物宽度通过以下公式计算：For example, the width of the obstacle at the lower edge of the target parking space is calculated by the following formula:

W2＝V*(T4-T3)W2=V*(T4-T3)

其中，W2表示目标车位下边缘的障碍物宽度，V表示车速，T3表示第三时刻，T4表示第四时刻。Among them, W2 represents the width of the obstacle at the lower edge of the target parking space, V represents the vehicle speed, T3 represents the third time, and T4 represents the fourth time.

S504.根据目标车位下边缘的障碍物宽度确定目标车位下边缘的补偿系数。S504. Determine the compensation coefficient of the lower edge of the target parking space according to the width of the obstacle at the lower edge of the target parking space.

目标车位下边缘的补偿系数的确定与目标车位上边缘的补偿系数的确定规则相同，在本步骤中，该补偿系数的设置与障碍物宽度成反比，即补偿系数的设置随障碍物宽度的增大而减小。The determination of the compensation coefficient of the lower edge of the target parking space is the same as the determination of the compensation coefficient of the upper edge of the target parking space. In this step, the setting of the compensation coefficient is inversely proportional to the width of the obstacle, that is, the setting of the compensation coefficient increases with the width of the obstacle Big and reduce.

S60.根据获取的超声波雷达测距数据计算目标车位下边缘的障碍物形状类型以及目标车位下边缘的补偿参数。S60. Calculate the obstacle shape type at the lower edge of the target parking space and the compensation parameter of the lower edge of the target parking space according to the acquired ultrasonic radar ranging data.

本步骤中，目标车位下边缘的障碍物形状类型的获取与目标车位上边缘的障碍物形状类型的获取方法相同，如图3所示，具体包括：In this step, the method for obtaining the shape type of the obstacle at the lower edge of the target parking space is the same as the method for obtaining the shape type of the obstacle at the upper edge of the target parking space, as shown in Figure 3, which specifically includes:

S601.获取多帧超声波雷达测距数据，并对超声波雷达测距数据进行滤波处理。S601. Acquire multiple frames of ultrasonic radar ranging data, and filter the ultrasonic radar ranging data.

此步骤中，多帧超声波雷达测距数据为第三时刻到第四时刻之间所获取的。In this step, the multi-frame ultrasonic radar ranging data is acquired between the third time and the fourth time.

S602.对处理后的超声波雷达测距数据进行缓存并计算每帧超声波雷达测距数据权重值。S602. Buffer the processed ultrasonic radar ranging data and calculate the weight value of each frame of ultrasonic radar ranging data.

S603.根据每帧超声波雷达测距数据权重值及连续获取的三帧超声波雷达测距数据之间的变化情况对缓存的超声波雷达测距数据进行处理得到有效超声波雷达测距数。S603. Process the buffered ultrasonic radar ranging data according to the weight value of each frame of ultrasonic radar ranging data and the changes between the three consecutive frames of ultrasonic radar ranging data to obtain the effective ultrasonic radar ranging number.

本步骤具体为：This step is specifically as follows:

S6031.分别判断每帧超声波雷达测距数据权重值是否大于第一预设阈值，若是，执行S6032，否则，执行S6035；S6031. Determine whether the weight value of each frame of ultrasonic radar ranging data is greater than the first preset threshold, if yes, execute S6032, otherwise, execute S6035;

S6032.判断连续获取的三帧超声波雷达测距数据是否超出超声波雷达测距范围，若是，执行S6035，否则，执行S6036；S6032. Determine whether the three consecutive frames of ultrasonic radar ranging data are beyond the ultrasonic radar ranging range, if yes, execute S6035, otherwise, execute S6036;

S6033.判断获取的当前帧超声波雷达测距数据是否大于上一帧超声波雷达测距数据，同时判断获取的当前帧超声波雷达测距数据是否大于下一帧超声波雷达测距数据，若是，执行S6035，否则，执行S6036；S6033. Determine whether the acquired current frame of ultrasonic radar ranging data is greater than the previous frame of ultrasonic radar ranging data, and determine whether the acquired current frame of ultrasonic radar ranging data is greater than the next frame of ultrasonic radar ranging data, if yes, execute S6035, Otherwise, execute S6036;

S6034.判断获取的当前帧超声波雷达测距数据是否小于上一帧超声波雷达测距数据，同时判断获取的当前帧超声波雷达测距数据是否小于下一帧超声波雷达测距数据；若是，执行S6035，否则，执行S6036；S6034. Determine whether the acquired current frame of ultrasonic radar ranging data is less than the previous frame of ultrasonic radar ranging data, and at the same time determine whether the acquired current frame of ultrasonic radar ranging data is less than the next frame of ultrasonic radar ranging data; if yes, execute S6035, Otherwise, execute S6036;

S6035.从缓存的超声波雷达测距数据中剔除该帧超声波雷达测距数据；S6035. Remove the frame of ultrasonic radar ranging data from the cached ultrasonic radar ranging data;

S6036.保留当前帧超声波雷达测距数据，从而得到有效超声波雷达测距数据。S6036. Keep the ultrasonic radar ranging data of the current frame, so as to obtain effective ultrasonic radar ranging data.

在本步骤中，第一预设阈值的设置范围为0.3-0.7，优选的第一预设阈值设置为0.5；同时，本步骤中还包括，当当前帧超声波雷达测距数据大于等于上一帧超声波雷达测距数据，同时当前帧超声波雷达测距数据小于等于下一帧超声波雷达测距数据，或者，当前帧超声波雷达测距数据小于等于上一帧超声波雷达测距数据，同时当前帧超声波雷达测距数据大于等于下一帧超声波雷达测距数据时，保留当前帧超声波雷达测距数据，从而得到有效超声波雷达测距数据。In this step, the setting range of the first preset threshold is 0.3-0.7, and the preferred first preset threshold is set to 0.5; at the same time, this step also includes, when the current frame of ultrasonic radar ranging data is greater than or equal to the previous frame Ultrasonic radar ranging data, while the current frame of ultrasonic radar ranging data is less than or equal to the next frame of ultrasonic radar ranging data, or the current frame of ultrasonic radar ranging data is less than or equal to the previous frame of ultrasonic radar ranging data, and the current frame of ultrasonic radar When the ranging data is greater than or equal to the next frame of ultrasonic radar ranging data, the current frame of ultrasonic radar ranging data is retained to obtain effective ultrasonic radar ranging data.

S604.提取有效超声波雷达测距数据所构建的障碍物轮廓特征，从而得到目标车位下边缘的障碍物形状类型，同时输出该障碍物形状类型。S604. Extracting the obstacle contour feature constructed by the effective ultrasonic radar ranging data, thereby obtaining the obstacle shape type at the lower edge of the target parking space, and outputting the obstacle shape type at the same time.

本步骤中，由于障碍物形状类型包括：方形、圆形和未知形。In this step, the types of obstacle shapes include: square, round, and unknown.

因此，此步骤具体为：提取有效超声波雷达测距数据所构建的障碍物轮廓特征，并对其进行分析，初始化该障碍物轮廓特征所构建的形状类型为未知形，然后判断该障碍物形状类型是否为方形，若是，则输出该障碍物形状类型，否则，再次判断障碍物形状类型是否为圆形，若是，则输出该障碍物形状类型，否则，判断障碍物形状类型为未知形并输出。Therefore, this step is specifically: extracting the obstacle contour feature constructed by the effective ultrasonic radar ranging data, and analyzing it, initializing the shape type constructed by the obstacle contour feature as an unknown shape, and then judging the obstacle shape type Whether it is square, if yes, output the obstacle shape type; otherwise, judge whether the obstacle shape type is circular again, if yes, output the obstacle shape type; otherwise, judge the obstacle shape type as unknown and output.

在平行车位中，目标车位的上边缘和下边缘所存在的障碍物仅为车辆的前端和后端，而车辆的前端和后端呈方形或圆形，因此当识别出的障碍物形状类型不是方形或圆形时，则表示目标车位的上边缘和下边缘所存在的障碍物不是车辆，对此不进行再次识别，直接输出其为未知形。In parallel parking spaces, the obstacles on the upper and lower edges of the target parking space are only the front and rear ends of the vehicle, while the front and rear ends of the vehicle are square or round. Therefore, when the obstacle shape type identified is not When it is square or round, it means that the obstacles on the upper and lower edges of the target parking space are not vehicles, and this is not recognized again, and it is directly output as an unknown shape.

S605.根据目标车位下边缘的障碍物的形状类型确定目标车位下边缘的补偿参数。S605. Determine the compensation parameter of the lower edge of the target parking space according to the shape type of the obstacle at the lower edge of the target parking space.

该目标车位下边缘的补偿参数的确定与目标车位上边缘的补偿参数的确定方式相同，在本步骤中，该补偿参数的设置与车速和障碍物形状类型所对应的边缘曲率呈正比，即补偿参数的设置随车速的增大以及障碍物形状类型所对应的边缘曲率的增大而增大。The determination of the compensation parameters of the lower edge of the target parking space is the same as the determination of the compensation parameters of the upper edge of the target parking space. In this step, the setting of the compensation parameter is proportional to the vehicle speed and the edge curvature corresponding to the obstacle shape type, that is, compensation The parameter setting increases with the increase of the vehicle speed and the increase of the edge curvature corresponding to the obstacle shape type.

S70.确定目标车位下边缘的补偿值。S70. Determine the compensation value of the lower edge of the target parking space.

本步骤中，目标车位下边缘的补偿值为目标车位下边缘补偿系数与目标车位下边缘补偿参数的乘积。In this step, the compensation value of the lower edge of the target parking space is the product of the lower edge compensation coefficient of the target parking space and the lower edge compensation parameter of the target parking space.

S80.计算目标车位上边缘与目标车位下边缘之间的距离。S80. Calculate the distance between the upper edge of the target parking space and the lower edge of the target parking space.

本步骤中，目标车位上边缘与目标车位下边缘之间的距离为：In this step, the distance between the upper edge of the target parking space and the lower edge of the target parking space is:

车辆在超声波雷达识别到目标车位下边缘的障碍物的起始边缘的第三时刻与超声波雷达识别到目标车位上边缘的障碍物的终止边缘的第二时刻之间的行驶位移。The driving displacement of the vehicle between the third time when the ultrasonic radar recognizes the starting edge of the obstacle on the lower edge of the target parking space and the second time when the ultrasonic radar recognizes the end edge of the obstacle on the upper edge of the target parking space.

目标车位上边缘与目标车位下边缘之间的距离的通过以下公式计算：The distance between the upper edge of the target parking space and the lower edge of the target parking space is calculated by the following formula:

L＝V*(T3-T2)L=V*(T3-T2)

其中，L表示目标车位上边缘与目标车位下边缘之间的距离，V表示车速，T3表示第三时刻，T2表示第二时刻。Among them, L represents the distance between the upper edge of the target parking space and the lower edge of the target parking space, V represents the vehicle speed, T3 represents the third time, and T2 represents the second time.

本步骤计算目标车位上边缘与目标车位下边缘之间的距离，可在识别出目标车位下边缘的障碍物的形状类型后计算，也可在识别出目标车位下边缘的障碍物宽度后计算，同样，还可在仅识别出目标车位下边缘的障碍物起始边缘时计算；This step calculates the distance between the upper edge of the target parking space and the lower edge of the target parking space. It can be calculated after identifying the shape type of the obstacle at the lower edge of the target parking space, or after identifying the obstacle width at the lower edge of the target parking space. Similarly, it can also be calculated when only the starting edge of the obstacle at the lower edge of the target parking space is recognized;

同时，目标车位上边缘与目标车位下边缘之间的距离可通过车速与超声波雷达识别到目标车位下边缘的障碍物的起始边缘的第三时刻与超声波雷达识别到目标车位上边缘的障碍物的终止边缘的第二时刻时差乘积计算，还可通过超声波雷达识别到目标车位下边缘的障碍物的起始边缘的测距数据与超声波雷达识别到目标车位上边缘的障碍物的终止边缘的测距数据之间的间隔计算。At the same time, the distance between the upper edge of the target parking space and the lower edge of the target parking space can be determined by the vehicle speed and the third time when the ultrasonic radar recognizes the obstacle on the lower edge of the target parking space and the ultrasonic radar recognizes the obstacle on the upper edge of the target parking space. Calculate the time difference product at the second time of the end edge of the target parking space. The distance measurement data of the starting edge of the obstacle at the lower edge of the target parking space can also be detected by the ultrasonic radar and the end edge measurement of the obstacle at the upper edge of the target parking space recognized by the ultrasonic radar. Calculate the interval between the data.

S90.计算目标车位长度。S90. Calculate the length of the target parking space.

在本步骤中，目标车位的长度根据目标车位上边缘与目标车位下边缘之间的距离、目标车位上边缘的补偿值和目标车位下边缘的补偿值计算。In this step, the length of the target parking space is calculated according to the distance between the upper edge of the target parking space and the lower edge of the target parking space, the compensation value of the upper edge of the target parking space, and the compensation value of the lower edge of the target parking space.

具体包括：目标车位长度为目标车位上边缘与目标车位下边缘之间的距离、目标车位上边缘的补偿值和目标车位下边缘的补偿值求和。Specifically, the length of the target parking space is the sum of the distance between the upper edge of the target parking space and the lower edge of the target parking space, the compensation value of the upper edge of the target parking space and the compensation value of the lower edge of the target parking space.

举例说明，目标车位的长度通过以下公式计算：For example, the length of the target parking space is calculated by the following formula:

SlotLen＝L+K1*Kh+K2*KtSlotLen=L+K1*Kh+K2*Kt

其中，SlotLen表示目标车位长度，L表示目标车位上边缘与目标车位下边缘之间的距离，K1表示目标车位上边缘的补偿系数，Kh表示目标车位上边缘的补偿参数，K1*Kh表示目标车位上边缘的补偿值，K2表示目标车位下边缘的补偿系数，Kt表示目标车位下边缘的补偿参数，K2*Kt表示目标车位下边缘的补偿值。Among them, SlotLen represents the length of the target parking space, L represents the distance between the upper edge of the target parking space and the lower edge of the target parking space, K1 represents the compensation coefficient of the upper edge of the target parking space, Kh represents the compensation parameter of the upper edge of the target parking space, K1*Kh represents the target parking space The compensation value of the upper edge, K2 represents the compensation coefficient of the lower edge of the target parking space, Kt represents the compensation parameter of the lower edge of the target parking space, and K2*Kt represents the compensation value of the lower edge of the target parking space.

显然，本发明的上述实施例仅仅是为清楚地说明本发明所作的举例，而并非是对本发明的实施方式的限定。对于所属领域的普通技术人员来说，在上述说明的基础上还可以做出其它不同形式的变化或变动。这里无需也无法对所有的实施方式予以穷举。凡在本发明的精神和原则之内所作的任何修改、等同替换和改进等，均应包含在本发明权利要求的保护范围之内。Obviously, the above-mentioned embodiments of the present invention are merely examples to clearly illustrate the present invention, and are not intended to limit the implementation of the present invention. For those of ordinary skill in the art, other changes or modifications in different forms can be made on the basis of the above description. It is unnecessary and impossible to list all the implementation methods here. Any modification, equivalent replacement and improvement made within the spirit and principle of the present invention shall be included in the protection scope of the claims of the present invention.

Claims

一种基于超声波传感器车位检测***的车位识别补偿方法，其特征在于，包括如下步骤：识别当前车速，并获取多帧超声波雷达测距数据；A parking space recognition and compensation method based on an ultrasonic sensor parking space detection system is characterized in that it comprises the following steps: recognizing the current vehicle speed and acquiring multiple frames of ultrasonic radar ranging data;

根据获取的超声波雷达测距数据分别计算目标车位上边缘和目标车位下边缘的障碍物宽度、目标车位上边缘与目标车位下边缘之间的距离；Calculate the obstacle width of the upper edge of the target parking space and the lower edge of the target parking space, and the distance between the upper edge of the target parking space and the lower edge of the target parking space according to the acquired ultrasonic radar ranging data;

根据获取的超声波雷达测距数据分别获取目标车位上边缘和目标车位下边缘的障碍物形状类型；Obtain the obstacle shape types at the upper edge of the target parking space and the lower edge of the target parking space according to the acquired ultrasonic radar ranging data;

根据车速、目标车位上边缘的障碍物宽度和目标车位上边缘的障碍物形状类型确定目标车位上边缘的补偿值；Determine the compensation value of the upper edge of the target parking space according to the vehicle speed, the width of the obstacle on the upper edge of the target parking space and the obstacle shape type on the upper edge of the target parking space;

根据车速、目标车位下边缘的障碍物宽度和目标车位下边缘的障碍物形状类型确定目标车位下边缘的补偿值；Determine the compensation value of the lower edge of the target parking space according to the vehicle speed, the width of the obstacle at the lower edge of the target parking space and the obstacle shape type at the lower edge of the target parking space;

根据目标车位上边缘与目标车位下边缘之间的距离、目标车位上边缘的补偿值和目标车位下边缘的补偿值计算目标车位长度。The length of the target parking space is calculated according to the distance between the upper edge of the target parking space and the lower edge of the target parking space, the compensation value of the upper edge of the target parking space and the compensation value of the lower edge of the target parking space.
根据权利要求1所述的基于超声波传感器车位检测***的车位识别补偿方法，其特征在于，障碍物形状类型的获取具体包括：The parking space recognition and compensation method based on the ultrasonic sensor parking detection system according to claim 1, wherein the obtaining of the obstacle shape type specifically includes:

获取多帧超声波雷达测距数据，并对超声波雷达测距数据进行滤波处理；Acquire multiple frames of ultrasonic radar ranging data, and filter the ultrasonic radar ranging data;

对处理后的超声波雷达测距数据进行缓存并计算每帧超声波雷达测距数据权重值；Cache the processed ultrasonic radar ranging data and calculate the weight value of each frame of ultrasonic radar ranging data;

根据每帧超声波雷达测距数据权重值及连续获取的三帧超声波雷达测距数据之间的变化情况对缓存的超声波雷达测距数据进行处理得到有效超声波雷达测距数据；According to the weight value of each frame of ultrasonic radar ranging data and the changes between the three consecutive frames of ultrasonic radar ranging data, the buffered ultrasonic radar ranging data is processed to obtain effective ultrasonic radar ranging data;

提取有效超声波雷达测距数据所构建的障碍物轮廓特征，从而得到障碍物形状类型。Extract the obstacle contour features constructed by the effective ultrasonic radar ranging data to obtain the obstacle shape type.
根据权利要求2所述的基于超声波传感器车位检测***的车位识别补偿方法，其特征在于，获取有效超声波雷达测距数据具体包括：The parking space recognition and compensation method based on the ultrasonic sensor parking detection system according to claim 2, wherein acquiring effective ultrasonic radar ranging data specifically includes:

分别判断每帧超声波雷达测距数据权重值是否大于第一预设阈值，若是，执行下一步，否则，从缓存的超声波雷达测距数据中剔除该帧超声波雷达测距数据；Respectively determine whether the weight value of each frame of ultrasonic radar ranging data is greater than the first preset threshold, if yes, proceed to the next step, otherwise, remove the frame of ultrasonic radar ranging data from the cached ultrasonic radar ranging data;

判断连续获取的三帧超声波雷达测距数据是否超出超声波雷达测距范围，若是，从缓存的超声波雷达测距数据中剔除该帧超声波雷达测距数据，否则，执行下一步；Determine whether the three consecutive frames of ultrasonic radar ranging data are beyond the ultrasonic radar ranging range, if so, remove the frame of ultrasonic radar ranging data from the cached ultrasonic radar ranging data, otherwise, proceed to the next step;

判断获取的当前帧超声波雷达测距数据是否大于上一帧超声波雷达测距数据，同时判断获取的当前帧超声波雷达测距数据是否大于下一帧超声波雷达测距数据，若是，从缓存的超声波雷达测距数据中剔除该帧超声波雷达测距数据，否则，执行下一步；Determine whether the acquired current frame of ultrasonic radar ranging data is greater than the previous frame of ultrasonic radar ranging data, and at the same time determine whether the acquired current frame of ultrasonic radar ranging data is greater than the next frame of ultrasonic radar ranging data, if so, from the cached ultrasonic radar Eliminate this frame of ultrasonic radar ranging data from the ranging data, otherwise, proceed to the next step;

判断获取的当前帧超声波雷达测距数据是否小于上一帧超声波雷达测距数据，同时判断获取的当前帧超声波雷达测距数据是否小于下一帧超声波雷达测距数据，若是，从缓存的超声波雷达测距数据中剔除该帧超声波雷达测距数据，否则，执行下一步；Determine whether the acquired current frame of ultrasonic radar ranging data is less than the previous frame of ultrasonic radar ranging data, and at the same time determine whether the acquired current frame of ultrasonic radar ranging data is less than the next frame of ultrasonic radar ranging data, if so, from the cached ultrasonic radar Eliminate this frame of ultrasonic radar ranging data from the ranging data, otherwise, proceed to the next step;

保留当前帧超声波雷达测距数据，从而得到有效超声波雷达测距数据。The ultrasonic radar ranging data of the current frame is retained to obtain effective ultrasonic radar ranging data.
根据权利要求3所述的基于超声波传感器车位检测***的车位识别补偿方法，其特征在于，在计算每帧超声波雷达测距数据的权重值之前，需先判断所缓存的超声波雷达测距数据的数量是否大于第二预设阈值，若是，计算每帧超声波雷达测距数据的权重值，否则，重新获取超声波雷达测距数据。The parking space recognition and compensation method based on the ultrasonic sensor parking detection system according to claim 3, characterized in that, before calculating the weight value of each frame of ultrasonic radar ranging data, it is necessary to first determine the number of ultrasonic radar ranging data buffered Whether it is greater than the second preset threshold, if so, calculate the weight value of each frame of ultrasonic radar ranging data; otherwise, re-acquire the ultrasonic radar ranging data.
根据权利要求4所述的基于超声波传感器车位检测***的车位识别补偿方法，其特征在于，所述第一预设阈值的设置范围为0.3-0.7，所述第二预设阈值的设置范围为3-7。The parking space recognition and compensation method based on the ultrasonic sensor parking detection system according to claim 4, wherein the setting range of the first preset threshold is 0.3-0.7, and the setting range of the second preset threshold is 3 -7.
根据权利要求2所述的基于超声波传感器车位检测***的车位识别补偿方法，其特征在于，每帧超声波雷达测距数据权重值的计算具体包括：The parking space recognition and compensation method based on the ultrasonic sensor parking detection system according to claim 2, wherein the calculation of the weight value of each frame of ultrasonic radar ranging data specifically includes:

缓存的超声波雷达测距数据中最大权重值和每帧超声波雷达测距数据的探测次数与测距概率因数的乘积之比。The ratio of the maximum weight value in the buffered ultrasonic radar ranging data to the product of the number of detections per frame of ultrasonic radar ranging data and the ranging probability factor.
根据权利要求2所述的基于超声波传感器车位检测***的车位识别补偿方法，其特征在于，障碍物形状类型包括：方形、圆形和未知形。The parking space recognition and compensation method based on the ultrasonic sensor parking detection system according to claim 2, wherein the obstacle shape types include: square, round, and unknown.
根据权利要求1所述的基于超声波传感器车位检测***的车位识别补偿方法，其特征在于，障碍物宽度的获取具体包括：The parking space recognition and compensation method based on the ultrasonic sensor parking detection system according to claim 1, wherein the obtaining of the width of the obstacle specifically includes:

识别障碍物起始边缘并记录当前时刻为第一时刻；Identify the starting edge of the obstacle and record the current moment as the first moment;

识别障碍物终止边缘并记录当前时刻为第二时刻；Identify the end edge of the obstacle and record the current moment as the second moment;

计算车辆在第一时刻到第二时刻之间的行驶距离为障碍物宽度。Calculate the distance of the vehicle from the first moment to the second moment as the width of the obstacle.
根据权利要求8所述的基于超声波传感器车位检测***的车位识别补偿方法，其特征在于，还包括：8. The parking space recognition and compensation method based on the ultrasonic sensor parking detection system according to claim 8, characterized in that it further comprises:

识别障碍物起始边缘，判断当前帧所获取的超声波雷达测距数据与上一帧所获取的超声波雷达测距数据之间的变化值是否大于第三预设阈值，若是，记录当前时刻为第一时刻，否则，重新识别障碍物起始边缘；Identify the starting edge of the obstacle, and determine whether the change value between the ultrasonic radar ranging data acquired in the current frame and the ultrasonic radar ranging data acquired in the previous frame is greater than the third preset threshold. If so, record the current moment as the first For a moment, otherwise, re-identify the starting edge of the obstacle;

识别障碍物终止边缘，判断当前帧所获取的超声波雷达测距数据与上一帧所获取的超声波雷达测距数据之间的变化值是否大于第三预设阈值，若是，记录当前时刻为第二时刻，否则，重新识别障碍物终止边缘。Identify the end edge of the obstacle, determine whether the change value between the ultrasonic radar ranging data obtained in the current frame and the ultrasonic radar ranging data obtained in the previous frame is greater than the third preset threshold, if so, record the current moment as the second At any time, otherwise, re-identify the obstacle termination edge.
根据权利要求9所述的基于超声波传感器车位检测***的车位识别补偿方法，其特征在于，所述第三预设阈值根据车辆宽度设置。The parking space recognition and compensation method based on the ultrasonic sensor parking space detection system according to claim 9, wherein the third preset threshold is set according to the width of the vehicle.
根据权利要求1所述的基于超声波传感器车位检测***的车位识别补偿方法，其特征在于，确定目标车位上边缘的补偿值和目标车位下边缘的补偿值具体包括：The parking space recognition and compensation method based on the ultrasonic sensor parking detection system according to claim 1, wherein determining the compensation value of the upper edge of the target parking space and the compensation value of the lower edge of the target parking space specifically comprises:

根据目标车位上边缘的障碍物宽度确定目标车位上边缘补偿系数，根据车速和目标车位上边缘的障碍物形状类型确定目标车位上边缘补偿参数，目标车位上边缘的补偿值为目标车位上边缘补偿系数与目标车位上边缘补偿参数的乘积；Determine the upper edge compensation coefficient of the target parking space according to the obstacle width of the upper edge of the target parking space, and determine the upper edge compensation parameters of the target parking space according to the vehicle speed and the obstacle shape type of the upper edge of the target parking space. The compensation value of the upper edge of the target parking space is the upper edge compensation of the target parking space The product of the coefficient and the edge compensation parameter of the target parking space;

根据目标车位下边缘的障碍物宽度确定目标车位下边缘补偿系数，根据车速和目标车位下边缘的障碍物形状类型确定目标车位下边缘补偿参数，目标车位下边缘的补偿值为目标车位下边缘补偿系数与目标车位下边缘补偿参数的乘积。Determine the lower edge compensation coefficient of the target parking space according to the obstacle width of the lower edge of the target parking space, and determine the lower edge compensation parameters of the target parking space according to the vehicle speed and the obstacle shape type of the lower edge of the target parking space. The compensation value of the lower edge of the target parking space is the bottom edge compensation of the target parking space. The product of the coefficient and the bottom edge compensation parameter of the target parking space.
根据权利要求11所述的基于超声波传感器车位检测***的车位识别补偿方法，其特征在于，目标车位上边缘补偿系数和目标车位下边缘补偿系数的获取具体包括：The parking space recognition and compensation method based on the ultrasonic sensor parking space detection system according to claim 11, wherein the obtaining of the upper edge compensation coefficient of the target parking space and the lower edge compensation coefficient of the target parking space specifically comprises:

补偿系数的设置随障碍物宽度的增大而减小。The setting of the compensation coefficient decreases as the width of the obstacle increases.
根据权利要求11所述的基于超声波传感器车位检测***的车位识别补偿方法，其特征在于，目标车位上边缘补偿参数和目标车位下边缘补偿参数的获取具体包括：The parking space recognition and compensation method based on the ultrasonic sensor parking space detection system according to claim 11, wherein the acquisition of the upper edge compensation parameter of the target parking space and the lower edge compensation parameter of the target parking space specifically comprises:

补偿参数的设置随车速的增大以及障碍物形状类型所对应的边缘曲率的增大而增大。The setting of the compensation parameter increases with the increase of the vehicle speed and the increase of the edge curvature corresponding to the obstacle shape type.
根据权利要求1所述的基于超声波传感器车位检测***的车位识别补偿方法，其特征在于，目标车位长度通过以下方式计算：The parking space recognition and compensation method based on the ultrasonic sensor parking detection system according to claim 1, wherein the length of the target parking space is calculated in the following manner:

目标车位长度为目标车位上边缘与目标车位下边缘之间的距离、目标车位上边缘的补偿值和目标车位下边缘的补偿值求和。The length of the target parking space is the sum of the distance between the upper edge of the target parking space and the lower edge of the target parking space, the compensation value of the upper edge of the target parking space and the compensation value of the lower edge of the target parking space.