A roadside perception equipment testing and evaluation method

Abstract

The present invention relates to the field of roadside sensing equipment testing, and in particular to a method for testing and evaluating roadside sensing equipment. The present invention proposes a method for testing and evaluating roadside sensing equipment, which enriches the current performance testing of roadside sensing equipment, which lacks a roadside equipment interference testing method and evaluation system based on real application scenarios under multi-lane and high-flow conditions, especially on expressways with 10 or more lanes in both directions and a daily traffic volume of more than 150,000 vehicles. It can accurately and effectively test the impact of roadside noise screens, roadside signs, gantry signs, streetlight poles and other roadside facilities on the detection and tracking effects of roadside sensing equipment. It improves the efficiency of the test while saving the test construction costs.

Description

Technical Field

[0001] The present invention relates to the field of roadside perception equipment testing, and in particular to a roadside perception equipment testing and evaluation method. Background Art

[0002] Roadside perception equipment is widely used in the field of intelligent transportation, and the test results of its equipment performance play an important reference role in equipment selection.

[0003] In the current performance testing of roadside sensing equipment, an ideal testing environment without any interference is often built, with only one or several motor vehicles slowly driving through the testing area. The roadside sensing equipment testing accuracy results obtained by this testing method cannot serve as a guide in actual projects. Especially in the case of ultra-wide sections and ultra-multiple lanes (i.e., 10 lanes or more in both directions) and ultra-large traffic (i.e., highways with daily traffic volume of more than 150,000 vehicles), the existing testing methods are difficult to accurately evaluate.

[0004] Therefore, based on actual engineering needs, the present invention proposes a roadside perception equipment testing and evaluation method. Summary of the invention

[0005] The purpose of the present invention is to overcome the problem that the existing testing and evaluation methods in the prior art are not practical, and to provide a roadside perception equipment testing and evaluation method.

[0006] In order to achieve the above-mentioned object of the invention, the present invention provides the following technical solutions:

[0007] A method for testing and evaluating a roadside sensing device comprises the following steps:

[0008] S1: Select a test data collection location from a preset location;

[0009] The test data collection location includes the relative distance between the roadside sensing device to be tested and the roadside interference facilities, the layout location of the roadside sensing device to be tested, and the installation height of the roadside sensing device to be tested;

[0010] S2: placing the roadside sensing device and the camera to be tested at the test data collection position, and adjusting the installation angles of the roadside sensing device and the camera to be tested;

[0011] S3: the roadside sensing device records detection data, and the camera records image data;

[0012] S4: selecting an unused test data collection position from the preset positions, and entering S2 until all the test data collection positions in the preset positions have completed recording;

[0013] S5: Based on the acquired detection data and image data of each test data collection position, the test results of the roadside equipment at each test point are evaluated, and the evaluation results are output as test results of the impact of roadside interference facilities on the detection and tracking effect of the roadside perception equipment.

[0014] As a preferred solution of the present invention, S2 adjusts the installation angle of the test data acquisition by a remote control pan / tilt mounted on a lifting vehicle, and the adjustment of the test data acquisition position is achieved by moving the lifting vehicle.

[0015] As a preferred solution of the present invention, the roadside sensing device is a radar-vision integrated device or a millimeter-wave radar;

[0016] The roadside interference facilities include: roadside noise screens and / or roadside signs and / or gantry signs and / or streetlight poles arranged along the road.

[0017] As a preferred embodiment of the present invention, S5 comprises the following steps:

[0018] According to the detection data of each test data collection position, the number of targets appearing in several sections in different lanes is counted to obtain the corresponding number of lane-by-lane target section detections of the roadside perception device;

[0019] According to the image data of each test data collection position, the number of targets appearing in several sections of different lanes is counted by the target detection algorithm to obtain the corresponding number of targets detected by the camera in each lane section;

[0020] Calculate the lane target detection rate; lane target detection rate = number of lane target sections detected by the roadside sensing device / number of lane target sections detected by the camera;

[0021] Draw a data heat map; the data heat map is drawn based on the detection data recorded by the roadside perception device. The more vehicles there are in a unit area during the statistical period, the darker the color of the heat map is, the better the continuous tracking of the target is, and the longer the continuous distance of the heat map is, which is used to qualitatively judge the perception situation of the perception device;

[0022] The number of lane-specific target section detections by the roadside perception device, the number of lane-specific target section detections by the camera, the lane-specific target detection rate, and the data heat map are output as test results of the impact of roadside interference facilities on the detection and tracking effects of roadside perception equipment.

[0023] As a preferred embodiment of the present invention, the method further comprises testing the road noise barrier, comprising the following steps:

[0024] Q1: Select the best acquisition position of the roadside sensing device to be tested based on the results output by S5, and use the other side of the highway noise barrier in the driving direction as the installation position of the roadside sensing device to be tested and the camera;

[0025] Q2: Set up the roadside sensing equipment and cameras to be tested;

[0026] Q3: The roadside sensing device to be tested records detection data, and the camera records image data;

[0027] Q4: Generate the number of lane-specific target section detections by the roadside perception device, the number of lane-specific target section detections by the camera, and a data heat map based on the detection data and image data recorded in S3 and Q3;

[0028] Q5: Calculate the lane target detection rate based on the number of lane target section detections by the roadside perception equipment and the number of lane target section detections by the camera generated in Q4;

[0029] Q6: The lane-by-lane target detection rate and data heat map output by S5, the lane-by-lane target detection rate output by Q5, and the data heat map output by Q4 are used as test results of the impact of highway noise barriers on the detection and tracking effects of the roadside perception equipment to be tested.

[0030] As a preferred solution of the present invention, in Q2, roadside sensing equipment and cameras are set up by laying out poles or lifting vehicles.

[0031] As a preferred embodiment of the present invention, the method further includes testing the cross-device continuous tracking capability under ultra-wide sections, ultra-multi-lanes, and ultra-large traffic, including the following steps:

[0032] E1: Select the optimal installation height of the roadside sensing device based on the output of S5;

[0033] E2: According to the output of Q6, the side that is less affected by the road noise barrier in the driving direction is selected as the installation side of the roadside sensing device;

[0034] E3: Three groups of roadside sensing devices and cameras are set at equal intervals at the optimal installation height on the installation side;

[0035] E4: In the order of driving direction, the collection direction of the two groups of roadside sensing devices and cameras in the front direction of the driving direction is set to the driving direction; the collection direction of the group of roadside sensing devices and cameras behind the driving direction is set to the opposite direction of driving;

[0036] E5: Select two groups of roadside perception devices and cameras with the same collection direction, record detection data and image data, obtain the number of lane-by-lane target matches in the detection overlap area, and calculate the lane-by-lane networking rate of the roadside perception devices in the same direction; the detection overlap area is the overlapping part of the corresponding target detection coverage of the two groups of roadside perception devices under test;

[0037] E6: Select two groups of roadside perception devices and cameras behind the driving direction, record detection data and image data, obtain the number of lane-by-lane target matches in the detection overlap area, and calculate the lane-by-lane networking rate of the reverse roadside perception devices;

[0038] E7: Based on the output results of E5 and E6, it is used as the test result of the cross-device continuous tracking capability under ultra-wide sections, ultra-multi-lanes, and ultra-large traffic.

[0039] As a preferred solution of the present invention, in E5 and E6, the lane-divided target matching number is the count value of the same target detected by two tested roadside sensing devices at the same time, at the same location, and in the same lane during the statistical period;

[0040] The calculation method of the lane-by-lane networking rate of the same-direction roadside perception device and the lane-by-lane networking rate of the opposite roadside perception device is: the ratio of the number of lane-by-lane target matches to the actual number of vehicles passing through the corresponding lane of the same section within the statistical period.

[0041] As a preferred solution of the present invention, the actual number of vehicles is obtained by analyzing the image data recorded by the camera through a video image algorithm.

[0042] A roadside perception device testing and evaluation device includes at least one processor and a memory communicatively connected to the at least one processor; the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute any of the methods described above.

[0043] Compared with the prior art, the present invention has the following beneficial effects:

[0044] The present invention proposes a roadside sensing equipment testing and evaluation method, which enriches the current lack of roadside sensing equipment performance testing under multi-lane and high-flow conditions, especially in highways with 10 or more lanes in both directions and a daily traffic volume of more than 150,000 vehicles. The roadside equipment interference testing method and evaluation system based on real application scenarios can accurately and effectively test the impact of roadside noise screens, roadside signs, gantry signs, streetlight poles and other roadside facilities on the detection and tracking effects of roadside sensing equipment. It saves test construction costs and improves test efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 This is a flow chart of a method for testing and evaluating a roadside sensing device according to Embodiment 1 of the present invention;

[0046] Figure 2 A schematic diagram of a test section in a method for testing and evaluating a roadside sensing device according to Embodiment 2 of the present invention;

[0047] Figure 3 A schematic diagram of an auxiliary test device in a roadside sensing device testing and evaluation method according to Embodiment 2 of the present invention;

[0048] Figure 4 A schematic diagram of lane detection rate in a roadside perception device testing and evaluation method according to Embodiment 2 of the present invention;

[0049] Figure 5 A heat map of the data of the integrated radar and vision device installed on the right side of the driving direction in a roadside perception device testing and evaluation method described in Example 3 of the present invention;

[0050] Figure 6 A heat map of the data of the integrated radar and vision device installed on the left side of the driving direction in a roadside perception device testing and evaluation method described in Example 3 of the present invention;

[0051] Figure 7 This is a schematic diagram of the layout of the integrated radar and vision devices when the illumination directions are the same in a roadside perception equipment testing and evaluation method described in Example 4 of the present invention;

[0052] Figure 8 This is a schematic diagram of the layout of the integrated radar and vision device when the illumination direction is opposite in a roadside perception device testing and evaluation method described in Example 4 of the present invention;

[0053] Fig. 9 This is a schematic structural diagram of a roadside perception device testing and evaluation device described in Example 5 of the present invention, which utilizes a roadside perception device testing and evaluation method of Example 1. DETAILED DESCRIPTION

[0054] The present invention is further described in detail below in conjunction with test examples and specific implementation methods. However, this should not be understood as the scope of the above subject matter of the present invention being limited to the following embodiments, and all technologies realized based on the content of the present invention belong to the scope of the present invention.

[0055] Example 1

[0056] like Figure 1 As shown, a roadside perception device testing and evaluation method includes the following steps:

[0057] S1: Select a test data collection location from a preset location;

[0058] The test data collection position includes the relative distance between the roadside sensing device to be tested and the roadside interference facilities, the layout position of the roadside sensing device to be tested, and the installation height of the roadside sensing device to be tested.

[0059] S2: Setting the roadside perception device and the camera to be tested at the test data collection position, and adjusting the installation angles of the roadside perception device and the camera to be tested.

[0060] S3: The roadside perception device records detection data, and the camera records image data.

[0061] S4: Select an unused test data collection position from the preset positions, and enter S2 until all the test data collection positions in the preset positions have completed recording.

[0062] S5: Based on the acquired detection data and image data of each test data collection position, the test results of the roadside equipment at each test point are evaluated, and the evaluation results are output as test results of the impact of roadside interference facilities on the detection and tracking effect of the roadside perception equipment.

[0063] Example 2

[0064] This embodiment is a specific implementation scheme of the method for testing and evaluating a roadside sensing device described in Embodiment 1, and includes the following steps:

[0065] S1: Select a test data collection location from a preset location;

[0066] The test data collection position includes the relative distance between the roadside sensing device to be tested and the roadside interference facilities, the layout position of the roadside sensing device to be tested, and the installation height of the roadside sensing device to be tested.

[0067] Among them, the roadside sensing equipment is a radar-vision integrated device or a millimeter-wave radar; the roadside interference facilities include: roadside noise screens and / or roadside signs and / or gantry signs and / or street light poles set up along the road.

[0068] Among them, Figure 2 As shown, this embodiment uses a certain section of highway as the test section, and selects the distance between the integrated radar and the gantry variable information board. In this embodiment, the preset positions include relative distances of 100 meters, 200 meters, and 300 meters; the installation heights are 6 meters, 8 meters, and 10 meters. Since the distance between each integrated radar and the gantry variable information board needs to be recorded and analyzed at different heights, the preset positions include 9 test data collection positions, and a total of 9 tests are required.

[0069] S2: Setting the roadside perception device and the camera to be tested at the test data collection position, and adjusting the installation angles of the roadside perception device and the camera to be tested.

[0070] In this embodiment, the test data collection position and the installation angle are adjusted by a remote control pan-tilt head arranged on the lifting vehicle, and the test data collection position is located above the emergency lane.

[0071] Further, such as Figure 3 As shown, this embodiment also includes an auxiliary testing device, which includes a lifting vehicle mounting bracket, a remote control pan-tilt head and a pan-tilt head mounting part, a radar-vision integrated machine mounting part, a variable focal length camera and a camera mounting part; the auxiliary testing device is installed on the lifting vehicle, and the radar-vision integrated machine, the remote control pan-tilt head and the variable focal length camera are installed on the auxiliary testing device.

[0072] In this embodiment, the lift truck is driven into the emergency lane, stopped at the selected position, the integrated radar and visual device to be tested is installed on the remote control pan-tilt, a variable focal length camera is installed, and the vehicle is lifted to the test installation height. The tester adjusts the installation angle of the integrated radar and visual device through the remote control pan-tilt until it meets the test requirements.

[0073] S3: The roadside sensing device records detection data, and the camera records image data. The recording time is generally controlled between 30 minutes and 1 hour, and the recording time in this embodiment is 30 minutes.

[0074] S4: Select an unused test data collection position from the preset positions, and enter S2 until all the test data collection positions in the preset positions have completed recording.

[0075] S5: Evaluate the test results of the roadside equipment at each test point based on the test data and image data obtained at each test data collection location, and output the evaluation results as test results of the impact of roadside interference facilities on the detection and tracking effect of roadside perception equipment. Specifically:

[0076] According to the detection data of each test data collection position, the number of targets appearing in several sections in different lanes is counted to obtain the corresponding number of lane-by-lane target section detections of the roadside perception equipment.

[0077] According to the image data of each test data collection position, the number of targets appearing in several sections of different lanes is counted through the target detection algorithm to obtain the corresponding number of targets detected by the camera in each lane section.

[0078] Calculate the lane target detection rate; lane target detection rate = number of lane target sections detected by the roadside perception equipment / number of lane target sections detected by the camera.

[0079] Draw a data heat map; the data heat map is drawn based on the detection data recorded by the roadside perception device. The more vehicles there are in a unit area during the statistical period, the darker the color of the heat map, the better the continuous tracking of the target, and the longer the continuous distance of the heat map, which is used to qualitatively judge the perception situation of the perception device.

[0080] The number of lane-specific target section detections by the roadside perception device, the number of lane-specific target section detections by the camera, the lane-specific target detection rate, and the data heat map are output as test results of the impact of roadside interference facilities on the detection and tracking effects of roadside perception equipment.

[0081] Furthermore, in this embodiment, the lanes are divided into lanes 1 to 5 to the left in the driving direction, and the cross-sections are divided from 100 meters to 350 meters at intervals of 50 meters.

[0082] According to the detection data of each test data collection position, the number of targets appearing in several sections in different lanes is counted to obtain the corresponding number of lane-by-lane target section detections of the roadside perception device;

[0083] According to the image data of each test data collection position, the number of targets appearing in several sections of different lanes is counted by the target detection algorithm to obtain the corresponding number of targets detected by the camera in each lane section;

[0084] Calculate the lane target detection rate; lane target detection rate = number of lane target sections detected by the roadside perception equipment / number of lane target sections detected by the camera.

[0085] In this embodiment, by analyzing the recorded data, the test part analysis results are obtained as follows:

[0086] Table 1 Test results of the first test unit

[0087]

[0088] like Figure 4 As shown in the figure, based on the analysis of the lane detection rate results, the lane detection rate of the integrated radar and vision device on a 5-lane road section with a 30-minute traffic volume of 2,618 motor vehicles decreases continuously with the increase of tracking distance.

[0089] Since the distance between the integrated radar and the gantry-type information board is 300 meters, when the tracking distance of the integrated radar and the gantry-type variable information board reaches 300 meters, its tracking rate shows an obvious turning point due to the influence of the gantry-type variable information board. Therefore, we can draw the following conclusions:

[0090] In this embodiment, the tracking rate of the integrated radar and vision device dropped significantly when it was 300 meters away from the gantry information board, indicating that the gantry information board had a greater impact on the tracking ability of the integrated radar and vision device at this distance. Moreover, on a 5-lane road with a 30-minute traffic volume of 2,618 vehicles, the tracking ability of the test integrated radar and vision device can still maintain a high level in lanes 1 and 2, but the tracking ability in lanes 3, 4, and 5 is poor.

[0091] Example 3

[0092] The difference between this embodiment and embodiment 2 is that the method further includes testing the road noise barrier, including the following steps:

[0093] Q1: Select the best acquisition position of the roadside perception device to be tested according to the results output by S5, and take the other side of the highway noise barrier in the driving direction as the installation position of the roadside perception device to be tested and the camera.

[0094] In this embodiment, the road noise barrier is arranged on the left side of the driving direction. Therefore, in order to test the influence of the road noise barrier on the detection and tracking effect of the integrated radar and vision device, the installation position of the integrated radar and vision device is selected to be on the right side of the driving direction.

[0095] According to the evaluation and analysis results of S5, the current test radar-vision integrated machine can obtain the best detection and tracking effect when it is installed at a height of 8 meters and 200 meters away from the gantry-type variable information board compared with other test schemes. Therefore, in this embodiment, the installation position of the radar-vision integrated machine is 200 meters away from the gantry-type variable information board and on a pole with a height of 8 meters.

[0096] Q2: Set up the roadside sensing equipment and cameras to be tested; set up the roadside sensing equipment and cameras by laying out poles or lifting vehicles.

[0097] Q3: The roadside perception device to be tested records detection data, and the camera records image data; the recording time of this embodiment is 30 minutes.

[0098] Q4: Based on the detection data and image data recorded in S3 and Q3, generate the number of lane-specific target section detections by the roadside perception device, the number of lane-specific target section detections by the camera, and a data heat map.

[0099] In this embodiment, the recorded data is analyzed to obtain a data heat map, such as Figure 5 and Figure 6 shown.

[0100] Q5: Calculate the lane target detection rate based on the number of lane target section detections by the roadside perception equipment and the number of lane target section detections by the camera generated in Q4.

[0101] Q6: The lane-by-lane target detection rate and data heat map output by S5, the lane-by-lane target detection rate output by Q5, and the data heat map output by Q4 are used as test results of the impact of highway noise barriers on the detection and tracking effects of the roadside perception equipment to be tested.

[0102] pass Figure 5 It can be seen that when a noise barrier is installed on the right side of the driving direction, the detection and tracking effect of the radar vision integrated device installed on the side close to the noise barrier is better than when it is installed on the side far away from the noise barrier.

[0103] Example 4

[0104] The difference between this embodiment and embodiment 3 is that the method further includes testing of cross-device continuous tracking capability under ultra-wide sections, ultra-multi-lanes, and ultra-large traffic, including the following steps:

[0105] E1: Select the optimal installation height of the roadside perception equipment according to the output result of S5; in this embodiment, the optimal installation height of the radar and vision integrated device is 8 meters.

[0106] E2: According to the result of Q6 output, select the side that is less affected by the highway noise barrier in the driving direction as the installation side of the roadside sensing device; further, when the radar-vision integrated device in this embodiment is installed on the side close to the noise barrier on a road covered by the noise barrier, the effect of the noise barrier is minimal. Therefore, in this embodiment, the radar-vision integrated device to be tested is installed on the side close to the noise barrier.

[0107] E3: Three groups of roadside sensing devices and cameras are arranged at equal intervals at the optimal installation height on the installation side; in this embodiment, the interval is set to 300 meters.

[0108] E4: In the order of driving direction, the collection direction of the two groups of roadside perception devices and cameras in the front of the driving direction is set to the driving direction; the collection direction of the group of roadside perception devices and cameras behind the driving direction is set to the opposite direction of driving.

[0109] E5: If Figure 7 As shown, two groups of roadside perception devices and cameras with the same collection direction are selected, detection data and image data are recorded (recorded for 1 hour in this embodiment), the number of lane target matches in the detection overlap area is obtained, and the lane networking rate of the roadside perception devices in the same direction is calculated; the detection overlap area is the overlapping part of the corresponding target detection coverage range of the two groups of roadside perception devices tested.

[0110] The lane target matching number is the count value of the same target detected by two tested roadside sensing devices at the same time, at the same location, and in the same lane during the statistical period.

[0111] The calculation method of the lane-by-lane networking rate of the same-direction roadside perception device and the lane-by-lane networking rate of the opposite roadside perception device is: the ratio of the number of lane-by-lane target matches to the actual number of vehicles passing through the corresponding lane of the same section within the statistical period.

[0112] The actual number of vehicles is obtained by analyzing the image data recorded by the camera through a video image algorithm.

[0113] Furthermore, the calculation results are as follows:

[0114] Table 2 Lane-based networking rate

[0115]

[0116] E6: If Figure 8 As shown, two groups of roadside perception devices and cameras behind the driving direction are selected, detection data and image data are recorded, the number of lane-by-lane target matches in the detection overlap area is obtained, and the lane-by-lane networking rate of the reverse roadside perception device is calculated.

[0117] Furthermore, the calculation results are as follows:

[0118] Table 3 Networking rate of integrated radar and vision devices by lane

[0119]

[0120] E7: Based on the output results of E5 and E6, it is used as the test result of the cross-device continuous tracking capability under ultra-wide sections, ultra-multi-lanes, and ultra-large traffic.

[0121] In this embodiment, according to the above calculation results, it can be known that when two integrated radar and vision devices are arranged in the order of driving direction and have the same illumination direction, the networking rate of the first two lanes close to the installation location of the integrated radar and vision device is maintained above 80%, and the networking rate of the 3rd, 4th and 5th lanes is significantly lower than that of the first two lanes.

[0122] When two integrated radar and vision devices are arranged in the order of driving direction and with opposite illumination directions, only the lane closest to the integrated radar and vision device can maintain a networking rate of more than 80%, and the networking rate of each lane is lower than that of the networking scheme in which the two integrated radar and vision devices have the same illumination direction.

[0123] Example 5

[0124] like Fig. 9 As shown, a roadside perception device testing and evaluation device includes at least one processor, a memory connected to the at least one processor, and at least one input / output interface connected to the at least one processor; the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute a roadside perception device testing and evaluation method of the aforementioned embodiment. The input / output interface may include a display, a keyboard, a mouse, and a USB interface for inputting and outputting data.

[0125] Those skilled in the art can understand that: all or part of the steps of implementing the above method embodiments can be completed by hardware related to program instructions, and the aforementioned program can be stored in a computer-readable storage medium. When the program is executed, it executes the steps of the above method embodiments; and the aforementioned storage medium includes: mobile storage devices, read-only memories (ROM), disks or optical disks, etc. Various media that can store program codes.

[0126] When the above-mentioned integrated unit of the present invention is implemented in the form of a software functional unit and sold or used as an independent product, it can also be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiment of the present invention can be essentially or partly reflected in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the methods of each embodiment of the present invention. The aforementioned storage medium includes: various media that can store program codes, such as mobile storage devices, ROMs, magnetic disks, or optical disks.

[0127] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A roadside perception device testing and evaluation method, characterized in that: The following steps are involved: S1: Select a test data collection location from a preset location; The test data collection location includes the relative distance between the roadside sensing device to be tested and the roadside interference facilities, the layout location of the roadside sensing device to be tested, and the installation height of the roadside sensing device to be tested; S2: placing the roadside sensing device and the camera to be tested at the test data collection position, and adjusting the installation angles of the roadside sensing device and the camera to be tested; S3: the roadside sensing device records detection data, and the camera records image data; S4: selecting an unused test data collection position from the preset positions, and entering S2 until all the test data collection positions in the preset positions have completed recording; S5: Evaluate the test results of the roadside equipment at each test point according to the acquired detection data and image data at each test data collection position, and output the evaluation results as test results of the influence of the roadside interference facilities on the detection and tracking effect of the roadside perception equipment; Further, S5 includes the following steps: According to the detection data of each test data collection position, the number of targets appearing in several sections in different lanes is counted to obtain the corresponding number of lane-by-lane target section detections of the roadside perception device; According to the image data of each test data collection position, the number of targets appearing in several sections of different lanes is counted by the target detection algorithm to obtain the corresponding number of targets detected by the camera in each lane section; Calculate the lane-by-lane target detection rate; Draw a data heat map; The number of lane-specific target section detections by the roadside perception device, the number of lane-specific target section detections by the camera, the lane-specific target detection rate, and the data heat map are output as test results of the impact of roadside interference facilities on the detection and tracking effects of roadside perception equipment.

2. A roadside perception equipment testing and evaluation method according to claim 1, characterized in that: S2 adjusts the installation angle of the test data acquisition by a remote control pan / tilt mounted on a lifting vehicle, and the adjustment of the test data acquisition position is achieved by moving the lifting vehicle.

3. A roadside perception equipment testing and evaluation method according to claim 1, characterized in that: The roadside sensing device is a radar-vision integrated device or a millimeter-wave radar; The roadside interference facilities include: roadside noise screens and / or roadside signs and / or gantry signs and / or streetlight poles arranged along the road.

4. A roadside perception equipment testing and evaluation method according to claim 1, characterized in that: The lane target detection rate = the number of lane target section detections by the roadside sensing device / the number of lane target section detections by the camera; the data heat map is drawn based on the detection data recorded by the roadside sensing device and is used to qualitatively judge the perception situation of the sensing device.

5. A roadside perception equipment testing and evaluation method according to claim 4, characterized in that: The method also includes testing the highway noise barrier, comprising the following steps: Q1: Select the best acquisition position of the roadside sensing device to be tested based on the results output by S5, and use the other side of the highway noise barrier in the driving direction as the installation position of the roadside sensing device to be tested and the camera; Q2: Set up the roadside sensing equipment and cameras to be tested; Q3: The roadside sensing device to be tested records detection data, and the camera records image data; Q4: Generate the number of lane-specific target section detections by the roadside perception device, the number of lane-specific target section detections by the camera, and a data heat map based on the detection data and image data recorded in S3 and Q3; Q5: Calculate the lane target detection rate based on the number of lane target section detections by the roadside perception equipment and the number of lane target section detections by the camera generated in Q4; Q6: The lane-by-lane target detection rate and data heat map output by S5, the lane-by-lane target detection rate output by Q5, and the data heat map output by Q4 are used as test results of the impact of highway noise barriers on the detection and tracking effects of the roadside perception equipment to be tested.

6. A roadside perception equipment testing and evaluation method according to claim 5, characterized in that: In Q2, roadside sensing equipment and cameras are installed by laying out poles or lifting vehicles.

7. A roadside perception equipment testing and evaluation method according to claim 5, characterized in that: The method also includes testing of cross-device continuous tracking capability under ultra-wide sections, ultra-multi-lanes, and ultra-large traffic, including the following steps: E1: Select the optimal installation height of the roadside sensing device based on the output of S5; E2: According to the output of Q6, the side that is less affected by the road noise barrier in the driving direction is selected as the installation side of the roadside sensing device; E3: Three groups of roadside sensing devices and cameras are set at equal intervals at the optimal installation height on the installation side; E4: In the order of driving direction, the collection direction of the two groups of roadside sensing devices and cameras in the front direction of the driving direction is set to the driving direction; the collection direction of the group of roadside sensing devices and cameras behind the driving direction is set to the opposite direction of driving; E5: Select two groups of roadside perception devices and cameras with the same collection direction, record detection data and image data, obtain the number of lane-by-lane target matches in the detection overlap area, and calculate the lane-by-lane networking rate of the roadside perception devices in the same direction; the detection overlap area is the overlapping part of the corresponding target detection coverage of the two groups of roadside perception devices under test; E6: Select two groups of roadside perception devices and cameras behind the driving direction, record detection data and image data, obtain the number of lane-by-lane target matches in the detection overlap area, and calculate the lane-by-lane networking rate of the reverse roadside perception devices; E7: Based on the output results of E5 and E6, it is used as the test result of the cross-device continuous tracking capability under ultra-wide sections, ultra-multi-lanes, and ultra-large traffic.

8. A roadside perception equipment testing and evaluation method according to claim 7, characterized in that: In E5 and E6, the lane-by-lane target matching number is the count value of the same target detected by the two tested roadside sensing devices at the same time, at the same location, and in the same lane during the statistical period; The calculation method of the lane-by-lane networking rate of the same-direction roadside perception device and the lane-by-lane networking rate of the opposite roadside perception device is: the ratio of the number of lane-by-lane target matches to the actual number of vehicles passing through the corresponding lane of the same section within the statistical period.

9. A roadside perception equipment testing and evaluation method according to claim 8, characterized in that: The actual number of vehicles is obtained by analyzing the image data recorded by the camera through a video image algorithm.

10. A roadside perception device testing and evaluation device, characterized in that: comprising at least one processor, and a memory communicatively connected to the at least one processor; The memory stores instructions that can be executed by at least one processor, and the instructions are executed by at least one processor to enable the at least one processor to perform the method according to any one of claims 1 to 9.

Images