Millimeter-wave radar model test and evaluation method based on simulation test platform
By obtaining the motion and radar measurement data of the target object in actual and simulated test scenarios, calculating various accuracy rates, and evaluating the detection performance of the vehicle-mounted millimeter-wave radar, the problem of ignoring the impact of target Doppler speed in the existing technology is solved, and the problem of ignoring the impact of target Doppler speed is improved. The accuracy of the evaluation was carried out.
Patent Information
- Application Number
- CN202411238341.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-05
- Publication Date
- 2024-11-08
- Estimated Expiration
- 2044-09-05
AI Technical Summary
The prior art is difficult to accurately evaluate the detection performance of vehicle-mounted millimeter-wave radar in complex environments in simulation testing, especially ignoring the impact of target Doppler speed on radar model detection accuracy.
By introducing target objects of different motion states in actual test scenarios and simulation test scenarios, real-time motion data and radar measurement data of target objects are obtained, the accuracy of target radial distance, Doppler velocity and horizontal azimuth angle is calculated, and The detection performance of the radar is evaluated based on these accuracy rates.
The real detection performance evaluation of millimeter-wave radar in complex environments is realized, which improves the accuracy of the evaluation and covers the static and dynamic motion states of the target object.
Smart Images

Figure CN118761239B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of vehicle-mounted radar simulation test, and in particular to a millimeter-wave radar model test and evaluation method based on a simulation test platform. Background Art
[0002] Autonomous driving technology has developed rapidly in recent years, and millimeter-wave radar has played an important role in realizing various autonomous driving functions. In order to experiment and test various autonomous driving algorithms in simulation, existing technologies often require the establishment of millimeter-wave radar models on autonomous driving simulation test platforms. At the same time, in order to ensure the performance of millimeter-wave radar models in simulation tests, it is also necessary to conduct relatively accurate evaluations of the models.
[0003] Related technology discloses a vehicle-mounted radar in-loop real-time simulation test system and method thereof, which includes: a radar under test on a turntable, a driving operation unit, a display, a real-time controller and a host computer; the driving operation unit includes: a driving operation mechanism and a driving operation signal sensor; the host computer includes: a virtual test field and a target test analysis unit; the virtual test field includes: a driving vehicle model and a target vehicle model; the driving vehicle model is used to virtually load the radar under test; the real-time controller includes: a spectrum measurement unit, a target simulation unit, a motion controller and a data acquisition card; the spectrum measurement unit includes: a millimeter wave signal transceiver, a millimeter wave downconverter, a millimeter wave signal analyzer, a delay controller, a millimeter wave signal generator, and a millimeter wave upconverter. This technology realizes the simulation test of the radar under various road conditions through radar hardware in-loop real-time simulation, driving operation, target simulation, turntable rotation control, and millimeter wave delay control.
[0004] However, in the process of implementing the present invention, the inventors found that the related art has the following problems:
[0005] The relevant technology tests the detection capability of the vehicle-mounted millimeter-wave radar in an ideal environment. In the actual driving environment, the static and dynamic elements in different scenes will also have a certain impact on the radar's target detection ability. In addition, the accuracy of radar detection can only be judged by the difference between the actual distance and azimuth of the target and the target distance and azimuth obtained by the simulation test, ignoring the influence of other important factors on the detection accuracy of the radar model, such as the target Doppler velocity. Summary of the invention
[0006] In order to solve the above problems existing in the prior art, the present invention provides a millimeter wave radar model test and evaluation method based on a simulation test platform, which is used to achieve accurate evaluation of the performance of the millimeter wave radar model. The technical solution provided by the present invention is as follows:
[0007] According to one aspect of an embodiment of the present invention, a millimeter wave radar model test and evaluation method based on a simulation test platform is provided, characterized in that the method comprises:
[0008] S10: constructing an actual test scene, wherein the actual test scene includes a road scene, a test vehicle located in the road scene, and various target objects, wherein the test vehicle is provided with a millimeter wave radar, and each target object moves according to a different preset trajectory;
[0009] S20: For each target object, the real movement data of the target object relative to the test vehicle within a preset time period is obtained in real time through the RT-Range system, and the real movement data includes a timestamp, a first radial distance corresponding to each timestamp, a first Doppler velocity, and a first horizontal azimuth; at the same time, the radar measurement data of the target object relative to the test vehicle within the preset time period is obtained in real time through the millimeter wave radar, and the radar measurement data includes a timestamp, a second radial distance corresponding to each timestamp, a second Doppler velocity, and a second horizontal azimuth;
[0010] S30: comparing the real motion data corresponding to the target object and the radar measurement data, and determining the radar measurement data that meets the preset accuracy as evaluation data;
[0011] S40: constructing a simulation test scene corresponding to the actual test scene, wherein the simulation test scene includes a virtual road scene, a simulation test vehicle located in the virtual road scene, and various simulation target objects, wherein the simulation test vehicle is provided with a simulated millimeter-wave radar to be tested, and each simulation target object moves according to the different preset trajectories;
[0012] S50: For each simulated target object, obtaining, in real time, simulated radar measurement data of the simulated target object relative to the simulated test vehicle within the preset time period through the simulated millimeter-wave radar to be tested, wherein the simulated radar measurement data includes a timestamp, a third radial distance corresponding to each timestamp, a third Doppler velocity, and a third horizontal azimuth;
[0013] S60: For each target object and the corresponding simulated target object, the accuracy of the target radial distance return value, the accuracy of the target horizontal azimuth return value, and the accuracy of the target Doppler velocity return value of the simulated millimeter-wave radar to be tested are respectively calculated according to the evaluation data and the corresponding simulated radar measurement data, and the detection performance of the simulated millimeter-wave radar to be tested is evaluated according to each accuracy rate.
[0014] Preferably, step S30 includes:
[0015] For the real motion data and radar measurement data corresponding to the target object, a radial distance difference between a first radial distance and a second radial distance, a Doppler velocity difference between a first Doppler velocity and a second Doppler velocity, and a horizontal azimuth angle difference between a first horizontal azimuth and a second horizontal azimuth are obtained at the same timestamp;
[0016] The radar measurement data whose radial distance difference satisfies the preset radial distance measurement accuracy, whose Doppler velocity difference satisfies the preset Doppler velocity measurement accuracy, and whose horizontal azimuth angle difference satisfies the preset horizontal azimuth angle measurement accuracy are determined as evaluation data.
[0017] Preferably, step S60 includes:
[0018] For each target object and the corresponding simulated target object, the radial distance difference between the second radial distance in the evaluation data corresponding to the target object and the third radial distance in the simulated radar measurement data corresponding to the simulated target object at the same timestamp is obtained, and the accuracy of the target radial distance return value of the simulated millimeter-wave radar to be tested is calculated according to Formula 1:
[0019] Formula 1
[0020] Wherein, A1 is the number of times that the radial distance difference meets the preset radial distance measurement accuracy within the preset time period, and A2 is the number of times that the radial distance difference does not meet the preset radial distance measurement accuracy within the preset time period. is the accuracy of the target radial distance return value of the simulated millimeter wave radar to be tested;
[0021] For each target object and the corresponding simulated target object, obtain the horizontal azimuth angle difference between the second horizontal azimuth in the evaluation data corresponding to the target object and the third horizontal azimuth in the simulated radar measurement data corresponding to the simulated target object at the same timestamp, and calculate the accuracy of the target horizontal azimuth return value of the simulated millimeter-wave radar to be tested according to Formula 2:
[0022] Formula 2
[0023] Wherein, B1 is the number of times that the radial distance difference meets the preset horizontal azimuth measurement accuracy within the preset time period, and B2 is the number of times that the radial distance difference does not meet the preset horizontal azimuth measurement accuracy within the preset time period. is the accuracy of the target horizontal azimuth angle return value of the simulated millimeter-wave radar to be tested;
[0024] For each target object and the corresponding simulated target object, obtain the Doppler velocity difference between the second Doppler velocity in the evaluation data corresponding to the target object and the third Doppler velocity in the simulated radar measurement data corresponding to the simulated target object at the same timestamp, and calculate the accuracy of the target Doppler velocity return value of the simulated millimeter-wave radar to be tested according to Formula 3:
[0025] Formula 3
[0026] Wherein, C1 is the number of times that the Doppler velocity difference meets the preset Doppler velocity measurement accuracy within the preset time period, and C2 is the number of times that the radial distance difference does not meet the preset Doppler velocity measurement accuracy within the preset time period. is the accuracy of the target Doppler velocity return value of the simulated millimeter-wave radar to be tested;
[0027] The performance evaluation value of the simulated millimeter-wave radar to be tested is obtained by multiplying each precision rate by the corresponding evaluation weight value, and the detection performance of the simulated millimeter-wave radar to be tested is determined and evaluated according to the performance evaluation value.
[0028] Preferably, the target objects include target vehicles and target pedestrians, and the motion state of each target object includes static and dynamic.
[0029] Preferably, the detection performance of the simulated millimeter-wave radar to be tested includes detection performance in a long-range mode and detection performance in a short-range mode. In this case, the method further includes:
[0030] After the millimeter-wave radar and the simulated millimeter-wave radar to be tested are respectively set to the long-range mode, steps S10 to S60 are performed to obtain a detection performance evaluation of the simulated millimeter-wave radar to be tested in the long-range mode;
[0031] After the millimeter-wave radar and the simulated millimeter-wave radar to be tested are respectively set to the short-range mode, steps S10 to S60 are performed to obtain a detection performance evaluation of the simulated millimeter-wave radar to be tested in the short-range mode.
[0032] Preferably, the method further comprises:
[0033] Different actual test scenarios are constructed to repeatedly execute steps S10 to S60, and the detection performance of the simulated millimeter-wave radar to be tested is comprehensively evaluated according to the accuracy of various detection performances of the simulated millimeter-wave radar to be tested in each actual test scenario.
[0034] Preferably, the simulation test scenario is constructed based on the PanoSim simulation test platform.
[0035] Preferably, the test vehicle and the corresponding simulation test vehicle are both in motion.
[0036] Compared with the prior art, the millimeter wave radar model test and evaluation method based on the simulation test platform provided by the present invention has the following advantages:
[0037] The present invention provides a millimeter-wave radar model testing and evaluation method based on a simulation test platform. By introducing target objects in different motion states into actual test scenes and corresponding simulation test scenes, and referring to the radial distance measurement accuracy, horizontal azimuth angle measurement accuracy and Doppler velocity measurement accuracy of the millimeter-wave radar between target objects, a comprehensive evaluation of the detection performance of the simulated millimeter-wave radar to be tested is performed, so that the test evaluation of the real detection performance of the simulated millimeter-wave radar to be tested in a complex environment is more accurate. BRIEF DESCRIPTION OF THE DRAWINGS
[0038] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.
[0039] Figure 1 The present invention is a flowchart of a method for testing and evaluating a millimeter-wave radar model based on a simulation test platform according to an exemplary embodiment of the present invention.
[0040] Figure 2 The figure is a calculation flow chart of evaluation data according to an exemplary embodiment of the present invention. DETAILED DESCRIPTION
[0041] In order to make the purpose, technical scheme and advantages of the present invention clearer, the present invention is described in detail below in combination with specific embodiments (but not limited to the embodiments) and the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0042] Figure 1 FIG. 1 is a flow chart of a method for testing and evaluating a millimeter wave radar model based on a simulation test platform according to an exemplary embodiment of the present invention. Figure 1 As shown, the millimeter wave radar model test and evaluation method based on the simulation test platform includes:
[0043] S10: constructing an actual test scene, wherein the actual test scene includes a road scene, a test vehicle located in the road scene, and various target objects, wherein the test vehicle is provided with a millimeter wave radar, and each target object moves according to a different preset trajectory.
[0044] The road scene includes a preset number of static elements, which are scene components that remain static in the road scene, including roads, traffic lights, traffic signs, lane lines, obstacles, sidewalks, and buildings beside the road.
[0045] Preferably, the target objects include target vehicles and target pedestrians, the motion state of each target object includes static and dynamic, and the target vehicles and target pedestrians are distributed around the test vehicle.
[0046] S20: For each target object, the real movement data of the target object relative to the test vehicle within a preset time period is obtained in real time through the RT-Range system, and the real movement data includes a timestamp, a first radial distance corresponding to each timestamp, a first Doppler velocity, and a first horizontal azimuth. At the same time, the radar measurement data of the target object relative to the test vehicle within the preset time period is obtained in real time through the millimeter-wave radar, and the radar measurement data includes a timestamp, a second radial distance corresponding to each timestamp, a second Doppler velocity, and a second horizontal azimuth.
[0047] When the target objects include template vehicles and target pedestrians, the test vehicle and the target vehicle are equipped with RT-Range systems, and the target pedestrian carries a backpack equipped with RT-Backpack accessories. The target vehicle and the target pedestrian move according to the preset trajectory within the detection range of the RT-Range system. The preset motion trajectory needs to include static, uniform motion, emergency stop and turning, etc. The motion direction needs to be away from the data collection vehicle, close to the data collection vehicle and perpendicular to the direction of the data collection vehicle, etc. Multiple sets of data are repeatedly collected, and the target motion trajectory of each set is different. The RT-Range system can track the target trajectory in real time, measure the radial distance, Doppler velocity and horizontal azimuth of the target object relative to the test vehicle, and save the relative radial distance, Doppler velocity and horizontal azimuth of the target object measured by the system as well as the timestamp of each frame of data as the real data of the motion. In addition, the corner reflectors are fixed at multiple positions in the scene, and the true value of the radial distance and horizontal azimuth of each corner reflector is recorded. Since the corner reflectors have a strong ability to reflect echo signals, the corner reflectors can be arranged at the boundary of the millimeter-wave radar detection range to verify the actual detection range of the millimeter-wave radar.
[0048] In the actual process of collecting real motion data, the installation positions of the RT-Range system components and the millimeter-wave radar on the test vehicle may be different. At this time, after obtaining the real motion data and radar measurement data of the test vehicle and the target object, the transformation matrix between the real motion data and the radar measurement data is calculated according to the actual position, and then the radar measurement data is transformed into the coordinate system of the RT-Range system for subsequent calculation and evaluation.
[0049] Exemplarily, partial real motion data of each test vehicle may be shown in Table 1.
[0050] Table 1
[0051]
[0052] S30: Compare the real motion data and radar measurement data corresponding to the target object, and determine the radar measurement data that meets the preset accuracy as evaluation data.
[0053] Preferably, the calculation flow chart of the evaluation data corresponding to step S30 can be as follows: Figure 2 As shown, including:
[0054] S31: For the real motion data and radar measurement data corresponding to the target object, obtain the radial distance difference between the first radial distance and the second radial distance, the Doppler velocity difference between the first Doppler velocity and the second Doppler velocity, and the horizontal azimuth angle difference between the first horizontal azimuth and the second horizontal azimuth at the same timestamp.
[0055] S32: Determine the radar measurement data whose radial distance difference satisfies the preset radial distance measurement accuracy, whose Doppler velocity difference satisfies the preset Doppler velocity measurement accuracy, and whose horizontal azimuth angle difference satisfies the preset horizontal azimuth angle measurement accuracy as evaluation data.
[0056] Since actual millimeter-wave radars have certain detection errors during the actual detection process, the embodiments of the present invention perform a certain degree of accuracy screening on radar measurement data to obtain screened evaluation data for subsequent simulation tests.
[0057] Among them, in a feasible embodiment, the measurement accuracy of each parameter corresponding to the millimeter wave radar can be as shown in Table 2.
[0058] Table 2
[0059]
[0060] S40: Construct a simulation test scene corresponding to the actual test scene, wherein the simulation test scene includes a virtual road scene, a simulation test vehicle located in the virtual road scene, and various simulation target objects, wherein the simulation test vehicle is provided with a simulated millimeter-wave radar to be tested, and each simulation target object moves according to the different preset trajectories.
[0061] Preferably, the simulation test scenario is constructed based on the PanoSim simulation test platform.
[0062] The WorldBuilder of the PanoSim simulation test platform can use actual road network data to build roads and arrange static elements such as traffic lights, traffic signs, obstacles, sidewalks and buildings beside the road in the scene. The dynamic elements are mainly meteorological and traffic participants. Meteorological conditions include weather and lighting conditions. Weather includes typical weather such as sunny, rainy, snowy and foggy. Lighting is composed of light intensity and sun altitude angle. WorldBuilder also supports the switching of typical weather and the modification of light intensity and sun altitude angle. Rainy and snowy weather can modify the rainfall / snow amount according to the actual scene, and foggy weather can modify the visibility of the scene.
[0063] The simulation test scene is constructed based on the simulation of the actual test scene, and each element in the actual test scene corresponds to each element in the simulation test scene. For example, the simulated millimeter-wave radar to be tested is arranged in the same position as the test vehicle in the actual scene. The simulated millimeter-wave radar to be tested in the simulation needs to have a consistent radar coordinate system with the actual millimeter-wave radar. If they are different, the coordinates of the simulated millimeter-wave radar to be tested need to be converted. The simulated millimeter-wave radar to be tested needs to have the same detection capability as the actual millimeter-wave radar. The parameters related to the detection capability of the simulated millimeter-wave radar to be tested include detection distance, horizontal field of view angle, vertical field of view angle, distance measurement accuracy, speed measurement accuracy, angle measurement accuracy, distance resolution, speed resolution and angle resolution.
[0064] S50: For each simulated target object, the simulated radar measurement data of the simulated target object relative to the simulated test vehicle within the preset time period is obtained in real time through the simulated millimeter-wave radar to be tested, and the simulated radar measurement data includes a timestamp, a third radial distance corresponding to each timestamp, a third Doppler velocity and a third horizontal azimuth.
[0065] S60: For each target object and the corresponding simulated target object, the accuracy of the target radial distance return value, the accuracy of the target horizontal azimuth return value, and the accuracy of the target Doppler velocity return value of the simulated millimeter-wave radar to be tested are respectively calculated according to the evaluation data and the corresponding simulated radar measurement data, and the detection performance of the simulated millimeter-wave radar to be tested is evaluated according to each accuracy rate.
[0066] The embodiment of the present invention evaluates the performance of the simulated millimeter-wave radar to be tested through the three dimensions of radial distance, horizontal azimuth angle, and Doppler velocity.
[0067] Preferably, step S60 includes:
[0068] S61: For each target object and the corresponding simulated target object, obtain the radial distance difference between the second radial distance in the evaluation data corresponding to the target object and the third radial distance in the simulated radar measurement data corresponding to the simulated target object at the same timestamp, and calculate the accuracy of the target radial distance return value of the simulated millimeter-wave radar to be tested according to Formula 1:
[0069] Formula 1
[0070] Wherein, A1 is the number of times that the radial distance difference meets the preset radial distance measurement accuracy within the preset time period, and A2 is the number of times that the radial distance difference does not meet the preset radial distance measurement accuracy within the preset time period. is the accuracy of the target radial distance return value of the simulated millimeter-wave radar to be tested.
[0071] For example, at a certain moment t The second radial distance returned by the actual scene acquisition is , the third radial distance return value in the simulation scene is , according to the running time statistics of each group of simulations and If the difference satisfies the radial distance measurement accuracy for 10,000 times and does not satisfy the radial distance measurement accuracy for 10,000 times, the accuracy of the target radial distance return value of the simulated millimeter-wave radar to be tested is calculated to be 50%.
[0072] S62: For each target object and the corresponding simulated target object, obtain the horizontal azimuth angle difference between the second horizontal azimuth in the evaluation data corresponding to the target object and the third horizontal azimuth in the simulated radar measurement data corresponding to the simulated target object at the same timestamp, and calculate the accuracy of the target horizontal azimuth return value of the simulated millimeter-wave radar to be tested according to Formula 2:
[0073] Formula 2
[0074] Wherein, B1 is the number of times that the radial distance difference meets the preset horizontal azimuth measurement accuracy within the preset time period, and B2 is the number of times that the radial distance difference does not meet the preset horizontal azimuth measurement accuracy within the preset time period. It is the accuracy of the target horizontal azimuth angle return value of the simulated millimeter-wave radar to be tested.
[0075] For example, at a certain moment t The second horizontal azimuth angle returned by the actual scene acquisition is , the return value of the third horizontal azimuth in the simulation scene is , according to the running time statistics of each group of simulations and The difference meets the horizontal azimuth angle measurement accuracy 5000 times and does not meet the horizontal azimuth angle measurement accuracy 15000 times, then the accuracy of the target horizontal azimuth angle return value calculated by the simulated millimeter-wave radar to be tested is 25%.
[0076] S63: For each target object and the corresponding simulated target object, obtain the Doppler velocity difference between the second Doppler velocity in the evaluation data corresponding to the target object and the third Doppler velocity in the simulated radar measurement data corresponding to the simulated target object at the same timestamp, and calculate the accuracy of the target Doppler velocity return value of the simulated millimeter-wave radar to be tested according to Formula 3:
[0077] Formula 3
[0078] Wherein, C1 is the number of times that the Doppler velocity difference meets the preset Doppler velocity measurement accuracy within the preset time period, and C2 is the number of times that the radial distance difference does not meet the preset Doppler velocity measurement accuracy within the preset time period. It is the accuracy of the target Doppler velocity return value of the simulated millimeter-wave radar to be tested.
[0079] For example, at a certain moment t The second Doppler velocity return value collected in the actual scene , the third Doppler velocity return value in the simulation scene is , according to the running time statistics of each group of simulations and If the difference meets the Doppler velocity measurement accuracy of 15000 times and is greater than the Doppler velocity measurement accuracy of 5000 times, the accuracy of the target Doppler velocity return value of the simulated millimeter-wave radar to be tested is calculated to be 75%.
[0080] S64: Multiply each precision rate by the corresponding evaluation weight value to obtain a performance evaluation value of the simulated millimeter-wave radar to be tested, and determine the detection performance of the simulated millimeter-wave radar to be tested according to the performance evaluation value.
[0081] Preferably, the detection performance of the simulated millimeter-wave radar to be tested includes detection performance in a long-range mode and detection performance in a short-range mode. In this case, the method further includes:
[0082] After setting the millimeter-wave radar and the simulated millimeter-wave radar to be tested to the long-range mode respectively, step S10 to step S60 are executed to obtain the detection performance evaluation of the simulated millimeter-wave radar to be tested in the long-range mode; after setting the millimeter-wave radar and the simulated millimeter-wave radar to be tested to the short-range mode respectively, step S10 to step S60 are executed to obtain the detection performance evaluation of the simulated millimeter-wave radar to be tested in the short-range mode.
[0083] Optionally, if the millimeter-wave radar has other working modes, the detection performance of the simulated millimeter-wave radar to be tested in the corresponding working mode can also be accurately evaluated according to steps S10 to S60.
[0084] Preferably, the method further comprises:
[0085] Different actual test scenarios are constructed to repeatedly execute steps S10 to S60, and the detection performance of the simulated millimeter-wave radar to be tested is comprehensively evaluated according to the accuracy of various detection performances of the simulated millimeter-wave radar to be tested in each actual test scenario.
[0086] Preferably, the test vehicle and the corresponding simulation test vehicle are both in motion.
[0087] In summary, the present invention provides a millimeter-wave radar model testing and evaluation method based on a simulation test platform, which introduces target objects in different motion states into the actual test scene and the corresponding simulation test scene, and simultaneously refers to the millimeter-wave radar's radial distance measurement accuracy, horizontal azimuth angle measurement accuracy and Doppler velocity measurement accuracy between target objects to perform a comprehensive evaluation of the detection performance of the simulated millimeter-wave radar to be tested, thereby making the test evaluation of the actual detection performance of the simulated millimeter-wave radar to be tested in a complex environment more accurate.
[0088] Those skilled in the art will readily appreciate other embodiments of the present invention after considering the specification and practicing the invention herein. The present invention is intended to cover any variations, uses or adaptations of the present invention that follow the general principles of the present invention and include common knowledge or customary techniques in the art that are not disclosed by the present invention. It should be understood that the present invention is not limited to the precise structure that has been described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from the scope thereof.
Claims
1. A millimeter wave radar model test and evaluation method based on a simulation test platform, It is characterized in that The method comprises: S10: constructing an actual test scene, wherein the actual test scene includes a road scene, a test vehicle located in the road scene, and various target objects, wherein the test vehicle is provided with a millimeter wave radar, and each target object moves according to a different preset trajectory; S20: For each target object, the real movement data of the target object relative to the test vehicle within a preset time period is obtained in real time through the RT-Range system, and the real movement data includes a timestamp, a first radial distance corresponding to each timestamp, a first Doppler velocity, and a first horizontal azimuth; at the same time, the radar measurement data of the target object relative to the test vehicle within the preset time period is obtained in real time through the millimeter wave radar, and the radar measurement data includes a timestamp, a second radial distance corresponding to each timestamp, a second Doppler velocity, and a second horizontal azimuth; S30: comparing the real motion data corresponding to the target object and the radar measurement data, and determining the radar measurement data that meets the preset accuracy as evaluation data; S40: constructing a simulation test scene corresponding to the actual test scene, wherein the simulation test scene includes a virtual road scene, a simulation test vehicle located in the virtual road scene, and various simulation target objects, wherein the simulation test vehicle is provided with a simulated millimeter-wave radar to be tested, and each simulation target object moves according to the different preset trajectories; S50: For each simulated target object, obtaining, in real time, simulated radar measurement data of the simulated target object relative to the simulated test vehicle within the preset time period through the simulated millimeter-wave radar to be tested, wherein the simulated radar measurement data includes a timestamp, a third radial distance corresponding to each timestamp, a third Doppler velocity, and a third horizontal azimuth; S60: For each target object and the corresponding simulated target object, the accuracy of the target radial distance return value, the accuracy of the target horizontal azimuth return value, and the accuracy of the target Doppler velocity return value of the simulated millimeter-wave radar to be tested are respectively calculated according to the evaluation data and the corresponding simulated radar measurement data, and the detection performance of the simulated millimeter-wave radar to be tested is evaluated according to each accuracy rate.
2. The millimeter wave radar model test and evaluation method according to claim 1, It is characterized in that Step S30 includes: For the real motion data and radar measurement data corresponding to the target object, a radial distance difference between a first radial distance and a second radial distance, a Doppler velocity difference between a first Doppler velocity and a second Doppler velocity, and a horizontal azimuth angle difference between a first horizontal azimuth and a second horizontal azimuth are obtained at the same timestamp; The radar measurement data whose radial distance difference satisfies the preset radial distance measurement accuracy, whose Doppler velocity difference satisfies the preset Doppler velocity measurement accuracy, and whose horizontal azimuth angle difference satisfies the preset horizontal azimuth angle measurement accuracy are determined as evaluation data.
3. The millimeter wave radar model test and evaluation method according to claim 1, It is characterized in that Step S60 includes: For each target object and the corresponding simulated target object, the radial distance difference between the second radial distance in the evaluation data corresponding to the target object and the third radial distance in the simulated radar measurement data corresponding to the simulated target object at the same timestamp is obtained, and the accuracy of the target radial distance return value of the simulated millimeter-wave radar to be tested is calculated according to Formula 1: Formula 1 Wherein, A1 is the number of times that the radial distance difference meets the preset radial distance measurement accuracy within the preset time period, and A2 is the number of times that the radial distance difference does not meet the preset radial distance measurement accuracy within the preset time period. is the accuracy of the target radial distance return value of the simulated millimeter-wave radar to be tested; For each target object and the corresponding simulated target object, obtain the horizontal azimuth angle difference between the second horizontal azimuth in the evaluation data corresponding to the target object and the third horizontal azimuth in the simulated radar measurement data corresponding to the simulated target object at the same timestamp, and calculate the accuracy of the target horizontal azimuth return value of the simulated millimeter-wave radar to be tested according to Formula 2: Formula 2 Wherein, B1 is the number of times that the radial distance difference meets the preset horizontal azimuth measurement accuracy within the preset time period, and B2 is the number of times that the radial distance difference does not meet the preset horizontal azimuth measurement accuracy within the preset time period. is the accuracy of the target horizontal azimuth angle return value of the simulated millimeter-wave radar to be tested; For each target object and the corresponding simulated target object, obtain the Doppler velocity difference between the second Doppler velocity in the evaluation data corresponding to the target object and the third Doppler velocity in the simulated radar measurement data corresponding to the simulated target object at the same timestamp, and calculate the accuracy of the target Doppler velocity return value of the simulated millimeter-wave radar to be tested according to Formula 3: Formula 3 Wherein, C1 is the number of times that the Doppler velocity difference meets the preset Doppler velocity measurement accuracy within the preset time period, and C2 is the number of times that the radial distance difference does not meet the preset Doppler velocity measurement accuracy within the preset time period. is the accuracy of the target Doppler velocity return value of the simulated millimeter-wave radar to be tested; The performance evaluation value of the simulated millimeter-wave radar to be tested is obtained by multiplying each precision rate by the corresponding evaluation weight value, and the detection performance of the simulated millimeter-wave radar to be tested is determined and evaluated according to the performance evaluation value.
4. The millimeter wave radar model test and evaluation method according to claim 1, It is characterized in that The target objects include target vehicles and target pedestrians, and the motion state of each target object includes static and dynamic.
5. The millimeter wave radar model test and evaluation method according to claim 1, It is characterized in that The detection performance of the simulated millimeter-wave radar to be tested includes the detection performance in the long-range mode and the detection performance in the short-range mode. At this time, the method further includes: After the millimeter-wave radar and the simulated millimeter-wave radar to be tested are respectively set to the long-range mode, steps S10 to S60 are performed to obtain a detection performance evaluation of the simulated millimeter-wave radar to be tested in the long-range mode; After the millimeter-wave radar and the simulated millimeter-wave radar to be tested are respectively set to the short-range mode, steps S10 to S60 are performed to obtain a detection performance evaluation of the simulated millimeter-wave radar to be tested in the short-range mode.
6. The millimeter wave radar model test and evaluation method according to claim 1, It is characterized in that The method further comprises: Different actual test scenarios are constructed to repeatedly execute steps S10 to S60, and the detection performance of the simulated millimeter-wave radar to be tested is comprehensively evaluated according to the accuracy of various detection performances of the simulated millimeter-wave radar to be tested in each actual test scenario.
7. The millimeter wave radar model test and evaluation method according to claim 1, It is characterized in that The simulation test scenario is built based on the PanoSim simulation test platform.
8. The millimeter wave radar model test and evaluation method according to claim 1, It is characterized in that The test vehicle and the corresponding simulation test vehicle are both in motion.
