Method and device for optimizing capability of roadside sensing equipment

By dynamically adjusting the lidar perception range, the problem of low perception accuracy of traditional roadside perception systems under complex road types and high traffic flow is solved, improving the real-time and accuracy of the system, and reducing hardware costs and complexity.

CN119049282APending Publication Date: 2024-11-29CHINA ACADEMY OF INFORMATION & COMM
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411184381.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-27
Publication Date
2024-11-29

AI Technical Summary

Technical Problem

Traditional roadside perception systems are difficult to flexibly adjust the lidar perception range under complex road types and high traffic flow conditions, resulting in a decrease in perception accuracy and limited computing capabilities, which affects real-time and accuracy.

Method used

By obtaining the actual driving trajectory and road type of the vehicle, a laser perceived target distance and perception error data set is generated, the data cluster is divided using clustering algorithms, the perception range of the lidar is dynamically adjusted, the perception error is optimized, and different road conditions are adapted.

Benefits of technology

It improves the perception accuracy and real-time performance of the roadside perception system, reduces computing load, reduces hardware costs and system design complexity, and ensures efficient operation in complex traffic environments.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119049282A_ABST
    Figure CN119049282A_ABST
Patent Text Reader

Abstract

The invention discloses a roadside sensing capability optimization method. The method comprises the following steps: obtaining an actual driving track of a vehicle and a driving track sensed by roadside sensing equipment located at a central point of a road intersection; for a set intersection type, generating a laser sensing target distance and sensing error data set in an intersection ROI (Region of Interest) under the condition of a straight lane and / or a turning lane; wherein each group of data sets corresponds to one intersection and one lane; and in the multiple groups of data sets, determining a common laser sensing target distance for various intersections and lanes, and enabling a sensing error corresponding to the common laser sensing target distance to be smaller than a set threshold value. The invention further comprises a device for implementing the method. The problem that the optimized laser sensing distance is not easy to determine under the actual road intersection condition is solved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of intelligent transportation technology, and in particular to a method and device for optimizing the capabilities of roadside sensing equipment. Background Art

[0002] In intelligent transportation systems, roadside perception systems play a vital role. The roadside perception system, which consists of sensors installed on the roadside, such as lidar, cameras, inertial navigation systems and GNSS (Global Navigation Satellite System), can perceive and analyze traffic conditions in real time, and provide accurate location and movement information of traffic participants. The roadside perception system not only plays an important role in traffic management, but also provides necessary environmental perception support for intelligent driving vehicles. However, due to the significant differences in the impact of different road types (such as T-junctions and crossroads) and their corresponding straight roads and curves on perception accuracy, how to adjust the perception range of lidar according to specific conditions has become the key to improving the system's perception capabilities.

[0003] Traditional roadside perception systems usually set the perception range of the LiDAR based on a fixed distance circle. This method fails to fully consider the differences in road types and driving paths. For example, at T-junctions and crossroads, the driving paths and dynamic behaviors of vehicles are more complex, including straight driving, left turns, and right turns. LiDARs with fixed perception ranges are difficult to flexibly adapt to these changes, resulting in a significant decrease in perception accuracy. In addition, the processing of roadside obstacles is usually filtered by setting the size of stationary objects. Although the influence of most stationary obstacles can be filtered out, the amount of data processed is large, which poses a considerable challenge to the computing power of the roadside perception system. Since the perception range of the LiDAR is fixed, the roadside perception system needs to process all data within the entire range, which not only requires powerful computing power, but may also lead to a decrease in real-time performance, affecting the timeliness and accuracy of traffic information. Especially in the case of heavy traffic, the processing capacity of the system is more significantly affected. The high-load data processing requirements not only increase the hardware cost, but also increase the complexity of system design and maintenance. For example, at a T-junction, due to its special geometry and variable traffic flow, the positioning accuracy of the vehicle is easily affected by environmental factors. In addition, buildings, billboards, and other stationary obstacles not only block the sensor's line of sight, but also reflect laser signals, increasing the error of the perception system. These environmental factors make it impossible for traditional fixed perception range methods to effectively cope with them, further reducing the accuracy and reliability of the perception system. In these complex intersections, the driving paths of vehicles are more diverse and unpredictable. Driving modes such as straight driving, left turns, and right turns coexist in the same place, requiring the perception system to be able to flexibly adjust the perception range in different directions to accurately detect and track the dynamics of each vehicle. This diversity further increases the difficulty of designing the perception system. Summary of the invention

[0004] The present application proposes a method and device for optimizing roadside perception capability to solve the problem that it is difficult to determine the optimized laser perception distance under actual road intersection conditions.

[0005] The present invention provides a method for optimizing roadside perception capability, including the following steps:

[0006] Obtain the actual driving trajectory of the vehicle and the driving trajectory perceived by the roadside sensing device located at the center point of the road intersection;

[0007] For a set intersection type, a data set of laser-sensed target distance and perception error in an intersection ROI area under straight road and / or turning road conditions is generated; wherein each set of the data sets corresponds to an intersection and a lane;

[0008] In multiple sets of data sets, a common laser perception target distance for multiple intersections and lanes is determined, and the perception error corresponding to the common laser perception target distance is less than a set threshold.

[0009] In one embodiment of the present application, the set intersection type includes one or more of the following intersection types: T-intersection, crossroad, and intersection with more than 4 directions.

[0010] In one embodiment of the present application, the perception error includes one or more of the following: positioning error and heading angle error.

[0011] In one embodiment of the present application, the step of determining a common laser sensing target distance for multiple intersections and lanes further includes:

[0012] For each type of intersection and lane, an optimal laser-perceived target distance is determined in the data set; and the optimal laser-perceived target distances from multiple data sets are averaged.

[0013] In one embodiment of the present application, the step of determining a common laser sensing target distance for multiple intersections and lanes further includes:

[0014] For each intersection and lane, a minimum perception error is determined in the data set; in each data set, the deviation between the perception error corresponding to the common laser perception target distance and the minimum perception error in the data set is determined; and the sum of the deviations determined in multiple data sets is the smallest.

[0015] In one embodiment of the present application, determining the laser sensing target distance and sensing error data set further comprises the following steps:

[0016] The roadside sensing device is located at the center of the road intersection. Within the ROI area, the actual driving trajectory of the vehicle and the driving trajectory perceived by the roadside sensing system are determined, and the actual position of the vehicle and the corresponding positioning error are compared;

[0017] Use a clustering algorithm to divide the data set into K clusters so that each data point belongs to the nearest cluster center;

[0018] According to the distance from each cluster center to the intersection center and the average positioning error of this cluster, a data set of laser-perceived target distance and corresponding positioning error is obtained.

[0019] In one embodiment of the present application, determining the laser sensing target distance and sensing error data set further comprises the following steps:

[0020] The roadside sensing device is located at the center of the road intersection. In the ROI area, the actual driving trajectory of the vehicle and the driving trajectory perceived by the roadside sensing device are obtained, and the actual position of the vehicle and the corresponding heading angle error are compared;

[0021] Use a clustering algorithm to divide the data set into K clusters so that each data point belongs to the nearest cluster center;

[0022] According to the distance from the center point of each cluster to the center point of the intersection and the average heading angle error of this cluster, a data set of laser-perceived target distance and corresponding heading angle error is obtained.

[0023] In one embodiment of the present application, the perception error is the statistical average of the total number of samples of multiple traffic participants detected at all perception moments within a set time.

[0024] The embodiment of the present application further provides a roadside perception capability optimization device, which is used to implement the method described in any embodiment of the present application, including:

[0025] An acquisition module, used to acquire the actual driving trajectory of the vehicle and the driving trajectory perceived by the roadside sensing device;

[0026] A generation module, for generating a data set of laser sensing target distance and sensing error in an intersection ROI area under straight road and / or turning road conditions for a set intersection type; wherein each set of the data sets corresponds to an intersection and a lane;

[0027] The determination module is used to determine a common laser perception target distance for multiple intersections and lanes in multiple data sets, and the perception error corresponding to the common laser perception target distance is less than a set threshold.

[0028] At least one of the above technical solutions adopted in the embodiments of the present application can achieve the following beneficial effects:

[0029] The present invention proposes a method for selecting the perception distance of a roadside perception system laser radar based on the road type and the driving path, so as to solve the problem of low perception accuracy caused by the perception range and roadside obstacles in the prior art. For example, by collecting and analyzing the driving data of vehicles located at the intersection, including straight driving, left turn and right turn, at T-junctions and crossroads, the roadside perception system at the intersection obtains a relationship model between the perception range and the positioning error and the heading angle error, and dynamically adjusts the perception range of the laser radar based on this model to improve the detection accuracy of the roadside system for traffic participants. At the same time, in order to further reduce the impact of roadside obstacles, buildings, etc. on the ability of the laser radar to perceive traffic participants, the corresponding perception range is delineated according to the actual road conditions, and the participation of objects outside the range is filtered out, so as to further enhance the perception ability of the laser radar. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0031] Figure 1 A diagram of the traffic scene application environment provided in the embodiment of the present application;

[0032] Figure 2 This is a flow chart of an embodiment of the method for optimizing the roadside perception capability of the present application;

[0033] Figure 3 This is a flow chart of an embodiment of the method of the present application including a clustering process;

[0034] Figure 4 An example of LiDAR target perception distance at a crossroads;

[0035] Figure 5 This is an example of LiDAR target perception at a T-junction;

[0036] Figure 6 An embodiment for determining a distance of a target sensed by a common laser;

[0037] Figure 7 An embodiment of a roadside perception capability optimization device of the present application;

[0038] Figure 8 An electronic device embodiment for implementing the method of the present application. DETAILED DESCRIPTION

[0039] In order to make the purpose, technical solution and advantages of the present application clearer, the technical solution of the present application will be clearly and completely described below in combination with the specific embodiments of the present application and the corresponding drawings. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present application.

[0040] The technical solution of this application can dynamically adjust the perception range. A system is designed that can flexibly adjust the laser radar perception range according to the road type and vehicle driving path, so that it can maintain high perception accuracy at T-junctions, crossroads, straight roads and curves. By dynamically adjusting the perception range, the system can effectively respond to changing traffic conditions and improve perception accuracy.

[0041] This application improves perception accuracy and real-time performance: By optimizing the perception range and data processing algorithm of the lidar, the amount of irrelevant data collected and processed is reduced, thereby improving the real-time performance and accuracy of the system. In complex traffic flow conditions, the system can respond quickly to ensure the timeliness and accuracy of traffic information.

[0042] The technical solution of this application reduces the computing load. By optimizing the data processing flow and perception range, setting a more specific perception range according to road conditions, reducing the amount of data that the system needs to process, thereby reducing hardware costs and system design complexity. It improves data processing efficiency and reduces the system's dependence on high-load computing data volume, so that it can still maintain efficient operation under cost control.

[0043] Using the technology of this application, the roadside perception system can also automatically adjust the perception strategy according to real-time traffic information and environmental changes, adapt to the changing traffic environment, and ensure the efficient operation of the perception system.

[0044] The technical solutions provided by various embodiments of the present application are described in detail below in conjunction with the accompanying drawings.

[0045] Figure 1 A diagram of the traffic scene application environment provided by the embodiment of this application. The center point of the intersection in this patent refers to the location of the roadside sensing device on the road, which is generally a device pile placed at the center of the intersection. The roadside sensing system is also composed of radar, sensing equipment, communication equipment, etc., such as Figure 1 As shown, a roadside sensing device 103 including a laser radar and a camera is combined with a central pole 104 at a road intersection, and performs wired or wireless communication with a computer device 101 with a visualization function through a communication network 102. The laser radar scans and performs a 360-degree rotation scan with its own position as the center of the circle.

[0046] The roadside sensing device 104 may be a radar or / and a camera, for example, at least one of a millimeter wave radar, a laser radar, and a camera. The computer device 101 may be a roadside computing unit / terminal / edge server. Optionally, the computer device may also be a cloud server, a vehicle-mounted computing unit / terminal, etc. The embodiment of the present application does not specifically limit the type of the computer device.

[0047] This patent only considers how to determine the optimal target perception distance of the radar in the roadside perception system. This perception distance is set to reduce the computational complexity of perception, improve perception performance, and reduce perception errors.

[0048] Figure 2 This is a flow chart of an embodiment of the roadside perception capability optimization method of the present application. The present application embodiment provides a roadside perception capability optimization method, including the following steps 210 to 230:

[0049] Step 210: Acquire the actual driving trajectory of the vehicle and the driving trajectory sensed by the roadside sensing device located at the center point of the road intersection.

[0050] The roadside perception test data of real vehicles driving on straight roads and curves at T-junctions and crossroads were collected. The straight road and curve data were analyzed and processed separately to provide measurement indicators for subsequent work.

[0051] Set the ROI (region of interest), which is a closed loop formed by connecting the longitude and latitude points at each intersection. In this area, by analyzing and processing the lidar point cloud data, camera video data, inertial navigation and GNSS data during the test period, the actual driving trajectory of the true value vehicle and the driving trajectory of the true value vehicle in this area perceived by the roadside perception system are obtained.

[0052] Step 220: for the set intersection type, generate a data set of laser-sensed target distance and perception error in the intersection ROI area under straight road and / or turning road conditions; wherein each set of the data sets corresponds to an intersection and a lane.

[0053] In one embodiment of the present application, the set intersection type includes one or more of the following intersection types: T-intersection, crossroad.

[0054] In one embodiment of the present application, the perception error includes one or more of the following: positioning error and heading angle error.

[0055] Preferably, the perception error is the statistical average of the total number of samples of multiple traffic participants detected at all perception moments within a set time.

[0056] Step 220A, positioning error

[0057] Within the perception range and within a specific time, calculate the average absolute error between the position detection value and the true value of the center point of all traffic participants by the roadside perception system in the local plane coordinate system:

[0058]

[0059] x i : The two-dimensional plane position vector of the i-th sample in the system to be tested;

[0060] y i : The true position vector of the i-th sample in the two-dimensional plane;

[0061] N: the total number of samples of all traffic participants detected at all perception moments within a specific time;

[0062] Through this calculation method, the average absolute error of the position point positioning of the straight road driving data based on the true value vehicle and the test vehicle at two road conditions, T-intersection and crossroad, at different time sampling points within a specific time period was obtained.

[0063] Step 220B, heading angle error

[0064] Because the probability of traffic accidents is higher on curves than on straight roads, and the change in the driving heading angle has the greatest impact on the process of driving on curves.

[0065] Similarly, a perception area is set to filter out irrelevant data and improve the computing power of the system. Then, within a specific time period of the perception range, the heading angles of all traffic participants perceived by the roadside perception system are collected, that is, the clockwise angle between the movement direction of the traffic participants and the north direction. Then, the average absolute error of the heading angle values ​​of the true value vehicle and the test vehicle is calculated:

[0066]

[0067] θ i : The heading angle of the system under test in the i-th sample;

[0068] μ i : The true heading angle of the i-th sample;

[0069] N: the total number of samples of all traffic participants detected at all perception moments within a specific time;

[0070] Based on this formula, the average absolute error of the heading angle of the curve driving data of the true vehicle and the test vehicle at two road conditions, a T-intersection and a crossroad, within a specific time period and at different time sampling points is obtained.

[0071] In one embodiment of the present application, the perception error is a positioning error, and determining the laser perception target distance and the perception error data set further comprises the following steps:

[0072] The roadside sensing device is located at the center of the road intersection. Within the ROI area, the actual driving trajectory of the vehicle and the driving trajectory perceived by the roadside sensing system are determined, and the actual position of the vehicle and the corresponding positioning error are compared;

[0073] Use a clustering algorithm to divide the data set into K clusters so that each data point belongs to the nearest cluster center;

[0074] According to the distance from each cluster center to the intersection center and the average positioning error of this cluster, a data set of laser-perceived target distance and corresponding positioning error is obtained.

[0075] In one embodiment of the present application, the perception error is a heading angle error, and determining the laser perception target distance and the perception error data set further comprises the following steps:

[0076] The roadside sensing device is located at the center of the road intersection. In the ROI area, the actual driving trajectory of the vehicle and the driving trajectory perceived by the roadside sensing device are obtained, and the actual position of the vehicle and the corresponding heading angle error are compared;

[0077] Use a clustering algorithm to divide the data set into K clusters so that each data point belongs to the nearest cluster center;

[0078] According to the distance from the center point of each cluster to the center point of the intersection and the average heading angle error of this cluster, a data set of laser-perceived target distance and corresponding heading angle error is obtained.

[0079] Step 230: Determine a common laser sensing target distance for multiple intersections and lanes in multiple data sets, and the sensing error corresponding to the common laser sensing target distance is less than a set threshold. For example, step 230A or step 230B.

[0080] Step 230A. In one embodiment of the present application, the step of determining a common laser perception target distance for multiple intersections and lanes further includes: for each intersection and lane, determining an optimal laser perception target distance in the data set; and averaging the optimal laser perception target distances from multiple data sets.

[0081] Step 230B. In one embodiment of the present application, the step of determining a common laser perception target distance for multiple intersections and lanes further includes: for each intersection and lane, determining a minimum perception error in the data set; determining in each data set the deviation between the perception error corresponding to the common laser perception target distance and the minimum perception error in the data set; the sum of the deviations determined in multiple data sets is the smallest.

[0082] It should be noted that the execution subject of each step of the method provided in the embodiment of the present application can be the same device, or the method can be executed by different devices. For example, the execution subject of step 210 and step 220 can be device 1, and the execution subject of step 230 can be device 2; for another example, the execution subject of step 210 can be device 1, and the execution subject of steps 220 and 230 can be device 2; and so on.

[0083] Figure 3 The flowchart of the embodiment of the clustering process included in the method of this application is as follows. First, the average absolute error value of the positioning at the sampling time point is preprocessed. Since the vehicle's position in straight driving and the heading angle in the process of turning will be affected by many factors (such as other traffic participants in the driving process and the characteristic factors of the intersection itself, etc.), a threshold is set to filter out the sampling point data with large errors to improve the accuracy of subsequent data processing; then the clustering algorithm is used to analyze and process the data.

[0084] For a given data set: S = {S1, S2, ..., S n The clustering algorithm divides the data set into K clusters, and makes each data point belong to the nearest cluster center. Its objective function is expressed as minimizing the intra-cluster sum of squares SSE, and the SSE is minimized by iteratively updating the cluster center:

[0085]

[0086] represents the center point of the jth cluster;

[0087] C: represents the set of all clusters;

[0088] According to the clustering results, the mean absolute error is divided into different clusters. Therefore, the characteristics of the straight road driving positioning accuracy error in each cluster under the two road conditions are obtained.

[0089] Furthermore, according to the clustering results, the heading angle mean absolute error is divided into different clustering clusters. Therefore, the characteristics of the heading angle accuracy error of curve driving in each cluster under the two road conditions are obtained.

[0090] Then, modeling and analysis are performed based on the distance from the center point of each cluster to the center point of the intersection. The relationship between the distance to the center point of the intersection and its positioning error at T-intersections, crossroads, and straight roads can be obtained more accurately. The straight road lidar target perception distance r1 is obtained, and then it is used as a measurement indicator for the subsequent lidar data distance circle selection to improve the roadside system's perception of the environment.

[0091] Alternatively, by modeling and analyzing the distance from each cluster center point to the intersection center point, the relationship between the distance to the intersection center point and its heading angle error at T-intersections and crossroads can be obtained more accurately during turning. The turning road lidar target perception distance r2 is obtained, which is then used as another measurement indicator for the subsequent lidar data distance circle selection to improve the roadside system's perception of the environment.

[0092] It should be noted that the roadside sensing devices are placed on the railings at the center of the road. At the intersection, due to the obstruction of buildings or multipath effects, the reception of positioning signals such as GPS may be affected, increasing the positioning error. If the distance is far, the signal attenuation problem is involved. When the vehicle is driving, the farther away from the roadside sensing device, the greater the positioning error, and the closer to the roadside sensing, the smaller the positioning error. It is necessary to comprehensively consider the relationship between distance and positioning error to determine the best sensing range.

[0093] Figure 4 This is an example of the LiDAR target sensing distance at an intersection. The figure shows the straight road at an intersection. Assume that the five-pointed star symbol is the location of the roadside sensing device, that is, the center point of the road intersection. The triangle symbol is the sampling point during vehicle driving. There are many sampling points, which are not listed here one by one. The arrow is the vehicle's straight driving trajectory. The distance value from each triangle to the five-pointed star in the figure is the distance from the vehicle to the center point of the intersection during driving.

[0094] Because the roadside sensing equipment senses the moving vehicle below, it will obtain the latitude and longitude of the vehicle at each sampling point. When the vehicle is driving, the laser radar system installed in the vehicle will also obtain its own latitude and longitude values ​​at each sampling point.

[0095] In the case of the same sampling point, there are two sets of longitude and latitude values. The values ​​collected by the vehicle itself are the actual values, and the values ​​perceived by the roadside perception system can be understood as the measured values. Through these two sets of longitude and latitude values, the positioning error can be obtained (reference to the determination of the error in the geographic coordinates). The longitude and latitude of the center point of the road (the location of the roadside perception system) are known. The distance from each sampling point to the center point of the intersection can be obtained. Through the above process, the lidar target perception distance and its positioning error are obtained.

[0096] For ease of understanding, as shown in the figure, assuming that the error of 0.1m corresponds to a distance of 15m, 0.9m corresponds to a distance of 25m, ..., 0.3m corresponds to a distance of 70m, it can be determined that the optimal perception distance of the roadside perception system is 15m and 70m. Since the distance of 15m is too short, if this is set as the optimal laser radar target perception distance, although the calculation complexity is reduced, in actual applications, only vehicles within a range of 15m are sensed, which has no practical value. Therefore, the optimal laser radar target perception distance for detecting straight vehicles is determined to be 70m.

[0097] The straight road principle at a T-intersection is similar to that at a crossroads, so I will not repeat the description here.

[0098] Figure 5 Take the lidar target perception for example at a curve at a T-junction. Similarly, in the case of a curve, the method for determining the lidar target perception distance is the same as the method for determining the distance on a straight road. Here, when the roadside perception system perceives a moving vehicle, it also collects the heading angle of the vehicle during its driving process. At the same time, the lidar system installed in the vehicle will also obtain its own heading angle at each sampling point. The self-measurement is the actual value of the heading angle, and the roadside perception system is the measured value. Through these two sets of heading angles, the heading angle error can be obtained. This obtains the lidar target perception distance of each sampling point and its corresponding heading angle error.

[0099] like Figure 5 , assuming that the heading angle error of 1° corresponds to a distance of 10m, 9° corresponds to a distance of 30m, ..., 3° ​​corresponds to a distance of 65m, it can be determined that the optimal perception distance of the roadside perception system is 10m and 65m. Since the distance of 10m is too short, if this is set as the optimal laser radar target perception distance, although the calculation complexity is reduced, in actual applications, only vehicles within a range of 10m are sensed, which has no practical value. Therefore, the optimal laser radar target perception distance for detecting vehicles driving on curves is determined to be 65m.

[0100] Figure 6 This is an embodiment of determining the target perception distance of a common laser. As in the method of step 230A, all data of the T-intersection and the crossroads are integrated, including the data of the straight road and the turning road of each intersection, to obtain the optimal laser radar target perception distance R1 for the T-intersection type; the optimal laser radar target perception distance R2 for the crossroad type.

[0101] In roadside perception, when analyzing LiDAR data, all traffic participants within the circle are usually selected for perception analysis based on the range of the distance circle. Through the first two analyses, two measurement indicators were obtained and applied to the selection of LiDAR distance circles. Through the previous work, the straight-line driving and turning driving data of T-intersections and crossroads were analyzed respectively, and the characteristics of each item were obtained. Through these characteristics, the range threshold for selecting the best LiDAR target perception distance was set.

[0102] The specific method is:

[0103] First, it is necessary to determine whether the intersection is a T-shaped intersection or a cross-shaped intersection. Based on the characteristics of the intersection, a perception region (ROI) is set to filter out the influence of irrelevant data. This reduces the amount of calculation of the roadside perception system while improving the perception capability.

[0104] Combined with the previous analysis of positioning error and heading angle error, the first indicator for selecting the optimal laser radar target perception distance is determined according to the characteristics of T-junctions and crossroads, and the second indicator for selecting the optimal laser radar target perception distance is determined according to whether the road is a straight road or a turning road. Finally, based on the first two analyses, the shared optimal laser radar target perception distance is obtained.

[0105] exist Figures 4-5 In the example, in order to reduce the complexity of calculation, the average method is directly used to determine the optimal laser radar target perception distance of the roadside perception system (as mentioned above, this patent only considers the radar in the roadside perception system) at the intersection: (70+65) / 2=67.5m. Therefore, the perception distance of the radar in the roadside perception system of the intersection can be set to 67.5m, which reduces the calculation complexity while ensuring the perception performance and optimizes the perception time of the system.

[0106] It should be noted that the beneficial effect of the average value is that it reduces the technical complexity. However, it does not mean that the farther away from the roadside sensing device, the greater the positioning error, nor does it mean that the closer to the roadside sensing device, the smaller the positioning error. It can be seen that the relationship between the error and the distance is not a linear relationship. Therefore, the distance value after arithmetic averaging may also be a point where both positioning errors increase. Therefore, as an optional embodiment of the present application, it is also possible to pursue a point close to the optimal values ​​of multiple perception errors as described in step 230B, rather than a point where the positions of the optimal values ​​of multiple positioning errors are compromised as described in step 230A.

[0107] Figure 7 This is an embodiment of a roadside perception capability optimization device of the present application. The present application embodiment also proposes a roadside perception capability optimization device, which is used to implement the method described in any embodiment of the present application, including:

[0108] The acquisition module 51 is used to acquire the actual driving trajectory of the vehicle and the driving trajectory sensed by the roadside sensing device. Correspondingly, the function described in step 210 of the embodiment is implemented.

[0109] The generation module 52 is used to generate a data set of laser sensing target distance and sensing error in the intersection ROI area under the conditions of a straight road and / or a turning road for the set intersection type; wherein each set of the data set corresponds to an intersection and a lane. Correspondingly, the function described in step 220 of the embodiment is implemented.

[0110] The determination module 53 is used to determine a common laser sensing target distance for multiple intersections and lanes in multiple data sets, and the sensing error corresponding to the common laser sensing target distance is less than a set threshold. Correspondingly, the function described in step 230 of the embodiment is implemented.

[0111] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0112] Therefore, the present application also proposes a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the method described in any embodiment of the present application is implemented.

[0113] Furthermore, the present application also proposes an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method described in any embodiment of the present application when executing the computer program.

[0114] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0115] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0116] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0117] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory. The memory may include non-permanent storage in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. Memory is an example of a computer-readable medium.

[0118] Figure 8 600 is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. The electronic device 600 shown is only an example and should not bring any limitation to the functions and scope of use of the embodiment of the present application. It includes: one or more processors 620; a storage device 610, which is used to store one or more programs, and when the one or more programs are run by the one or more processors 620, the one or more processors 620 implement the steps of the method described in the embodiment of the present application.

[0119] The electronic device 600 further includes an input device 630 and an output device 640; the processor 620, storage device 610, input device 630 and output device 640 in the electronic device can be connected via a bus or other means, and the figure takes the connection via a bus 650 as an example.

[0120] The storage device 610, as a computer-readable storage medium, can be used to store software programs, computer executable programs, and module units, such as program instructions corresponding to the method in the embodiment of the present application. The storage device 610 may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application required for at least one function; the storage data area may store data created according to the use of the terminal, etc. In addition, the storage device 610 may include a high-speed random access memory, and may also include a non-volatile memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage device. In some instances, the storage device 610 may further include a memory remotely arranged relative to the processor 620, and these remote memories may be connected via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0121] The input device 630 may be used to receive input numbers, character information or voice information, and generate key signal input related to user settings and function control of the electronic device. The output device 640 may include electronic devices such as a display screen and a speaker.

[0122] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0123] The above is only an embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.

Claims

1. A method for optimizing roadside perception capability, characterized in that: The following steps are involved: Obtain the actual driving trajectory of the vehicle and the driving trajectory perceived by the roadside sensing device located at the center point of the road intersection; For a set intersection type, a data set of laser-sensed target distance and perception error in an intersection ROI area under straight road and / or turning road conditions is generated; wherein each set of the data sets corresponds to an intersection and a lane; In multiple sets of data sets, a common laser perception target distance for multiple intersections and lanes is determined, and the perception error corresponding to the common laser perception target distance is less than a set threshold.

2. A method for optimizing roadside perception capability according to claim 1, characterized in that: The set intersection type includes one or more of the following intersection types: T-intersection, crossroad; and / or, The perception error includes one or more of the following: positioning error and heading angle error.

3. The method for optimizing roadside perception capability according to claim 1, characterized in that: The step of determining a common laser sensing target distance for multiple intersections and lanes further includes: For each type of intersection and lane, an optimal laser-perceived target distance is determined in the data set; and the optimal laser-perceived target distances from multiple data sets are averaged.

4. The method for optimizing roadside perception capability according to claim 1, characterized in that: The step of determining a common laser sensing target distance for multiple intersections and lanes further includes: For each intersection and lane, a minimum perception error is determined in the data set; in each data set, the deviation between the perception error corresponding to the common laser perception target distance and the minimum perception error in the data set is determined; and the sum of the deviations determined in multiple data sets is the smallest.

5. The method for optimizing roadside perception capability according to claim 1, characterized in that: Determining the laser sensing target distance and sensing error data set further includes the following steps: The roadside sensing device is located at the center of the road intersection. Within the ROI area, the actual driving trajectory of the vehicle and the driving trajectory perceived by the roadside sensing system are determined, and the actual position of the vehicle and the corresponding positioning error are compared; Use a clustering algorithm to divide the data set into K clusters so that each data point belongs to the nearest cluster center; According to the distance from each cluster center to the intersection center and the average positioning error of this cluster, a data set of laser-perceived target distance and corresponding positioning error is obtained.

6. A method for optimizing roadside perception capability according to claim 1, characterized in that: Determining the laser sensing target distance and sensing error data set further includes the following steps: The roadside sensing device is located at the center of the road intersection. In the ROI area, the actual driving trajectory of the vehicle and the driving trajectory perceived by the roadside sensing device are obtained, and the actual position of the vehicle and the corresponding heading angle error are compared; Use a clustering algorithm to divide the data set into K clusters so that each data point belongs to the nearest cluster center; According to the distance from the center point of each cluster to the center point of the intersection and the average heading angle error of this cluster, a data set of laser-perceived target distance and corresponding heading angle error is obtained.

7. A method for optimizing roadside perception capability according to claim 1, characterized in that: The perception error is the statistical average of the total number of samples of multiple traffic participants detected at all perception moments within a set time.

8. A roadside perception capability optimization device, used to implement the method according to any one of claims 1 to 7, characterized in that: include: An acquisition module, used to acquire the actual driving trajectory of the vehicle and the driving trajectory perceived by the roadside sensing device; A generation module, for generating a data set of laser sensing target distance and sensing error in an intersection ROI area under straight road and / or turning road conditions for a set intersection type; wherein each set of the data sets corresponds to an intersection and a lane; The determination module is used to determine a common laser perception target distance for multiple intersections and lanes in multiple data sets, and the perception error corresponding to the common laser perception target distance is less than a set threshold.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

10. An electronic device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method according to any one of claims 1 to 7 is implemented.