Laser radar performance test method and device

Abstract

The invention provides a laser radar performance test method and device, and the method comprises the steps: setting a detection position, enabling the detection position to be located at the central position of a laser radar view field partition, and employing a laser radar to carry out the detection of a target statically disposed at the detection position, so as to determine the static point cloud data; driving the turntable to rotate at a preset angular speed so as to drive the target arranged at the edge of the turntable to perform circular motion at a preset speed, so that the motion path of the target passes through the detection position; detecting the motion state of the target through a time sequence synchronization module to determine motion time sequence data of the target; determining a synchronization time sequence signal based on the motion time sequence data of the target and the position information of the detection position; in response to the synchronous time sequence signal, performing time sequence synchronous data acquisition through a laser radar to obtain synchronous point cloud data; and determining dynamic performance parameters of the laser radar view field partition based on the static point cloud data and the synchronous point cloud data, and completing a laser radar performance test.

Description

Technical Field

[0001] This disclosure relates to the field of lidar detection and testing technology, and in particular to a lidar performance testing method and apparatus. Background Technology

[0002] Current dynamic performance testing of automotive LiDAR typically employs static target testing, evaluating radar performance by detecting point cloud density, range accuracy, and reflection intensity on a fixed reflective surface. However, this method fails to accurately reflect the LiDAR's response capability, range update rate, and signal echo dynamics when detecting high-speed moving targets (such as pedestrians, vehicles, and foreign objects). Traditional testing generally uses static targets (such as reflectors, calibration spheres, or checkerboard patterns) to measure ranging accuracy, repeatability, angular resolution, signal-to-noise ratio, reflectivity characteristics, and scan consistency. However, in this type of testing: the radar's scanning time delay (from a few milliseconds to tens of milliseconds) is ignored; system latency and data timestamp accuracy are not reflected; intra-frame distortion of moving targets is completely absent; and the spatiotemporal consistency of the point cloud cannot be verified. Therefore, even with very high static accuracy, it cannot represent the vehicle's perception performance of high-speed objects while in motion. Introducing dynamic target testing (i.e., the target or platform has a known speed and trajectory relative to the radar) can reflect the "motion error" in real-world usage scenarios. However, current dynamic target testing still suffers from poor accuracy and cumbersome testing processes.

[0003] Therefore, there is an urgent need for a method and device for testing the performance of lidar that can simulate real dynamic moving targets, so as to better test the performance of lidar under controllable conditions. Summary of the Invention

[0004] In view of the above, this disclosure provides a method and apparatus for testing the performance of lidar, the technical solution of which is as follows:

[0005] According to embodiments of this disclosure, a method for testing the performance of a lidar is provided, comprising: setting a detection position and adjusting the installation angle of the lidar mounted on a lidar fixing module, such that the detection position is located at the center of a lidar field of view partition, wherein the lidar field of view partition is one of multiple lidar field of view partitions determined by dividing the lidar field of view; using the lidar to detect a target statically set at the detection position to determine static point cloud data; driving a turntable to rotate at a preset angular velocity to drive a target set at the edge of the turntable to perform circular motion at a preset speed, wherein the motion path of the target passes through the detection position; detecting the motion state of the target through a timing synchronization module to determine the motion timing data of the target; determining a synchronization timing signal based on the motion timing data of the target and the position information of the detection position; responding to the synchronization timing signal, acquiring synchronization point cloud data by using the lidar to perform timing synchronization data acquisition; and determining the dynamic performance parameters of the lidar field of view partition based on the static point cloud data and the synchronization point cloud data, thereby completing the lidar performance test.

[0006] According to embodiments of this disclosure, determining the dynamic performance parameters of a lidar field of view partition based on static point cloud data and synchronous point cloud data includes: performing correction processing on the static point cloud data to obtain corrected static point cloud data; performing a fitting operation on the corrected static point cloud data to determine the reference center axis of the target; grouping the synchronous point cloud data to obtain N synchronous point cloud data groups, where N≥2; performing a fitting operation on each of the N synchronous point cloud data groups to determine N segmented center axes; and determining the dynamic performance parameters of the lidar field of view partition based on the reference center axis and the N segmented center axes.

[0007] According to an embodiment of this disclosure, the target motion timing data includes real-time target position data. Determining a synchronization timing signal based on the target motion timing data and the position information of the detection position includes: detecting the motion timing data, and generating a synchronization timing signal when the real-time target position data and the position information of the detection position are the same.

[0008] According to an embodiment of this disclosure, in response to a synchronization timing signal, synchronous point cloud data is obtained by acquiring timing synchronization data through a lidar, including: in response to the synchronization timing signal, a control module controls the lidar to acquire point cloud data to obtain intermediate point cloud frame data; and extracting the intermediate point cloud frame data to determine the target point cloud cluster, thereby obtaining synchronous point cloud data.

[0009] According to embodiments of this disclosure, the dynamic performance parameters of a lidar field of view partition include: tilt angle, distortion, and a comprehensive dynamic distortion index. The method for determining the dynamic performance parameters of a lidar field of view partition includes: determining the tilt angle based on a reference center axis and N segmented center axes; determining the N-1 segmented angles between adjacent segmented center axes based on the N segmented center axes; determining the distortion based on the N-1 segmented angles; and determining the comprehensive dynamic distortion index based on the tilt angle and distortion.

[0010] According to embodiments of this disclosure, when there are multiple targets, the method further includes: when acquiring static point cloud data, detecting each target separately to obtain multiple static point cloud data, and marking the multiple static point cloud data with corresponding target information; when acquiring synchronization timing signals, identifying each target passing through the detection position based on the target recognition module to obtain each target information, and marking the corresponding synchronization timing signals with the target information to obtain multi-target synchronization timing signals; when determining synchronization point cloud data, acquiring synchronization point cloud data in response to the multi-target synchronization timing signals, and marking the synchronization point cloud data based on the target information corresponding to the synchronization timing signals to obtain synchronization point cloud data for multiple targets; in the data processing stage, performing data analysis on static point cloud data and synchronization point cloud data with the same target information to determine multiple sets of dynamic performance parameters of the lidar for multiple targets.

[0011] According to embodiments of this disclosure, the lidar performance testing method further includes: determining a new preset speed by adjusting the turntable rotation speed and / or the distance between the target and the turntable rotation axis, and performing dynamic performance testing of the lidar at the new preset speed; and achieving dynamic performance testing of all lidar field-of-view zones by adjusting the installation angle of the lidar mounted on the lidar fixed module.

[0012] According to embodiments of this disclosure, a lidar performance testing device is also provided, including a lidar fixing module, a disc rotation mechanism, a timing synchronization module, and a control module. The radar fixing module includes a bracket and an adjustment platform. The bracket is configured to be fixed relative to the turntable position. The adjustment platform is configured to be fixedly connected to the bracket via a first end and to the lidar via a second end. By adjusting the tilt angle of the adjustment platform, the detection position is positioned at the center of the lidar's field of view partition. The lidar's field of view partition is one of multiple lidar field of view partitions defined by dividing the lidar's field of view. The disc rotation mechanism includes a turntable and a drive mechanism. The turntable is configured to drive a target positioned on the edge of the turntable to perform circular motion under the drive mechanism. The target's motion path passes through the detection position. The timing synchronization module is configured to detect the target's motion state to determine the target's motion timing data and determine a synchronization timing signal based on the target's motion timing data and the detection position's location information. The control module is communicatively connected to the disc rotation mechanism, the lidar, and the timing synchronization module. It is configured to acquire static point cloud data and synchronized point cloud data and determine the dynamic performance parameters of the lidar's field of view partition based on the static point cloud data and synchronized point cloud data.

[0013] According to an embodiment of this disclosure, the edge of the turntable is provided with at least one mounting hole for mounting at least one target; the control module includes a speed control unit, a static data acquisition unit, a synchronization timing determination unit, a synchronization point cloud data acquisition unit, and a performance analysis unit. The speed control unit controls the rotational speed of the turntable based on a communication connection with the turntable rotation mechanism; the static data acquisition unit obtains static point cloud data based on a communication connection with the lidar; the synchronization timing determination unit determines the synchronization timing signal based on a communication connection with the timing synchronization module; the synchronization point cloud data acquisition unit responds to the synchronization timing signal and controls the lidar to perform timing synchronization data acquisition to determine the synchronization point cloud data; the performance analysis unit determines the dynamic performance parameters of the lidar's field of view partition based on the static point cloud data and the synchronization point cloud data.

[0014] According to an embodiment of this disclosure, the timing synchronization module includes a rotation angle measuring unit and a synchronization signal generating unit disposed on the rotating shaft of the turntable; the rotation angle measuring unit is configured to collect and output real-time rotation angle data of the turntable; the synchronization signal generating unit is configured to receive the real-time rotation angle data and determine the synchronization timing signal based on the real-time rotation angle data and the detection position. Attached Figure Description

[0015] The above and other objects, features and advantages of this disclosure will become clearer from the following description of embodiments with reference to the accompanying drawings, in which:

[0016] Figure 1This is a flowchart illustrating the lidar performance testing method according to an embodiment of the present disclosure.

[0017] Figure 2 This is a schematic diagram of a lidar performance testing device according to an embodiment of the present disclosure. Detailed Implementation

[0018] This disclosure provides a method and apparatus for testing the performance of a lidar. By constructing a dynamic target simulation device with adjustable speed, it enables dynamic detection and testing of targets with different speeds and reflection characteristics, thereby realistically and quantitatively evaluating the dynamic response characteristics and other performance of the lidar. This disclosure simulates dynamic targets based on circular trajectories, allowing for a wider adjustable speed range and synchronization with the lidar's timing sequence.

[0019] When a vehicle is traveling at high speed or a target is approaching at high speed, the point cloud of a LiDAR (LiDAR) is not collected simultaneously, but scanned point by point. Dynamic target testing can quantify the deformation caused by scanning delay (e.g., the target "stretching" or "tilting" in the point cloud). This is especially important for multi-line mechanical radar (distortion is significant when the beams are not synchronized). It verifies time synchronization and delay performance. The timestamp accuracy of the output data, the overall system delay, and inter-frame consistency can be measured using dynamic targets (such as a constantly moving reflector). These directly affect sensor fusion and positioning accuracy in advanced driver assistance systems and autonomous driving. Therefore, this disclosure provides a test device capable of simulating real dynamic moving targets for testing the dynamic performance of LiDAR under controlled conditions, including indicators such as response delay, distance error, reflection stability, and moving target recognition capability.

[0020] For example, when a vehicle-mounted lidar scans a high-speed moving target, its point cloud acquisition is a sequential process performed line by line, resulting in millisecond-level differences in the measurement time of each scan line. When the target translates or rotates during the scan, the spatial position of the target corresponding to different scan lines changes, causing asynchronous distortion (rolling distortion) in the point cloud along the direction of motion. For targets with regular geometry that are always visible as a single reflection point during rotation (such as a corner bevel reflector array or a cone), this distortion manifests in the 3D point cloud as a velocity-dependent tilt and multi-segment distortion along the fitted central axis: when the target moves at a constant speed, the point cloud tilts as a whole along the scanning direction, with the tilt angle proportional to the target speed and the radar scan cycle; when the radar uses a multi-line scanning structure (such as multi-channel rotation or solid-state layered scanning), due to the fixed deviation in the sampling timing of different scanning layers, the point cloud segments acquired by each layer form "segmented center lines," which are no longer collinear in space and exhibit velocity-dependent segmented distortion. This composite distortion of tilt and twist is a direct geometric manifestation of temporal inconsistency during motion, and can sensitively reflect the degree of mismatch between radar ranging and time synchronization performance under high-speed conditions.

[0021] To make the objectives, technical solutions, and advantages of this disclosure clearer, the following detailed description is provided in conjunction with specific embodiments and the accompanying drawings.

[0022] In this disclosure, a method for testing the performance of a lidar is provided, such as... Figure 1 As shown, the test method includes the following steps:

[0023] S1: Set the detection position and adjust the installation angle of the lidar installed on the lidar fixed module so that the detection position is at the center of the lidar field of view partition. The lidar field of view partition is one of the multiple lidar field of view partitions determined by dividing the lidar field of view.

[0024] S2: Use lidar to detect a static target set at the detection location to determine static point cloud data;

[0025] S3: Drive the turntable to rotate at a preset angular velocity, so as to drive the target set on the edge of the turntable to perform circular motion at a preset speed, wherein the motion path of the target is determined by the detection position;

[0026] S4: The motion state of the target is detected by the timing synchronization module to determine the timing data of the target's motion;

[0027] S5 determines the synchronization timing signal based on the target's motion timing data and the detection location information;

[0028] S6: In response to the synchronization timing signal, synchronized point cloud data is obtained by acquiring timing synchronization data using lidar; and

[0029] S7: Determine the dynamic performance parameters of the lidar field of view partition based on static point cloud data and synchronous point cloud data, and complete the lidar performance test.

[0030] According to an embodiment of this disclosure, the radar fixing module includes a bracket and an adjustment platform. The lidar is mounted on the adjustment platform with an adjustable mounting angle. A detection position is set and positioned at the center of the lidar's field of view partition. The bracket is configured to be fixed relative to the turntable position and can be adjusted according to actual detection needs. The adjustment platform is configured to be fixedly connected to the bracket via a first end and to the lidar via a second end. For example, the detection position can be positioned at the center of the lidar's field of view partition by adjusting the tilt angle of the adjustment platform itself. The lidar field of view partition is one of multiple lidar field of view partitions determined by dividing the lidar's field of view. For example, the lidar field of view partition may include three field of view partitions: left, center, and right. Alternatively, each of the three field of view partitions can be further divided into three partitions: upper, middle, and lower, resulting in nine lidar field of view partitions.

[0031] According to an embodiment of this disclosure, a target is disposed on a turntable, which is configured to drive the target disposed on the edge of the turntable to perform circular motion under the drive of a drive mechanism, wherein the motion path of the target can be detected by position detection.

[0032] According to an embodiment of this disclosure, the target motion timing data includes real-time target position data. Determining a synchronization timing signal based on the target motion timing data and the position information of the detection position includes: detecting the motion timing data, and generating a synchronization timing signal when the real-time target position data and the position information of the detection position are the same.

[0033] According to an embodiment of this disclosure, step S6 includes:

[0034] In response to the synchronization timing signal, the control module controls the lidar to acquire point cloud data to obtain intermediate point cloud frame data; and extracts the intermediate point cloud frame data to determine the target point cloud cluster, thereby obtaining synchronized point cloud data.

[0035] According to an embodiment of this disclosure, step S7 includes:

[0036] S71: Perform correction processing on the static point cloud data to obtain corrected static point cloud data;

[0037] S72: Perform a fitting operation based on the corrected static point cloud data to determine the reference center axis of the target;

[0038] S73: Based on the synchronized point cloud data, data grouping is performed to obtain N synchronized point cloud data groups, where N≥2;

[0039] S74: Perform fitting operations on N groups of synchronized point cloud data to determine N segmented central axes; and

[0040] S75: Based on the reference center axis and N segmented center axes, determine the dynamic performance parameters of the lidar field of view partition.

[0041] According to embodiments of this disclosure, the dynamic performance parameters of a lidar field of view partition include: tilt angle, distortion, and a comprehensive dynamic distortion index. The method for determining the dynamic performance parameters of a lidar field of view partition includes: determining the tilt angle based on a reference center axis and N segmented center axes; determining the N-1 segmented angles between adjacent segmented center axes based on the N segmented center axes; determining the distortion based on the N-1 segmented angles; and determining the comprehensive dynamic distortion index based on the tilt angle and the distortion.

[0042] To quantitatively characterize this dynamic distortion, the technical solution disclosed herein is based on a controllable speed rotating target and a high-precision time synchronization module to realize the acquisition and analysis of dynamic point clouds.

[0043] Assuming the target is at rest under ideal conditions, the central axis of the cone or conical target is... The precise spatial equation is determined through multi-frame averaging or static scanning. Furthermore, synchronous point cloud data (i.e., dynamic point cloud data) is acquired: when the target moves at an angular velocity... or linear velocity During movement, record the radar point cloud and the timestamp corresponding to each point. Then, the central axis is fitted piecewise: the point cloud is grouped according to the scan channel or time period, and least-squares fitting is performed on each group of points to obtain the corresponding piecewise central axis. Calculate the velocity-related velt angle and twist index: velocity-related velt angle. The center axis of each segment and the reference center axis The average angle reflects the overall velocity-related distortion, then:

[0044]

[0045] in, This represents the offset along the Z-axis in the coordinate system of the lidar. This represents the offset on the X-axis in the coordinate system of the lidar. denoted as the single-frame scan time, and D as the target size.

[0046] Furthermore, calculate the degree of distortion. The root mean square of the angle between the direction vectors of the central axes of adjacent scan segments is used to characterize the non-collinearity of multiple segments.

[0047]

[0048] Where k represents the ordinal number.

[0049] Finally, a comprehensive distortion evaluation index is defined as follows: The comprehensive dynamic distortion index is: ,in These are weighting coefficients, which can be set according to the experimental objectives.

[0050] According to embodiments of this disclosure, when there are multiple targets, the lidar performance testing method further includes:

[0051] Each static target set at the detection location is detected separately to obtain multiple static point cloud data, and the multiple static point cloud data are marked with the corresponding target information respectively;

[0052] When acquiring the synchronization timing signal, the target recognition module identifies each target passing through the detection position to obtain the target information, and uses the target information to mark the corresponding synchronization timing signal to obtain the multi-target synchronization timing signal.

[0053] When determining the synchronized point cloud data, synchronized point cloud data is acquired in response to the synchronization timing signal of multiple targets. Based on the target information corresponding to the synchronization timing signal, the synchronized point cloud data is labeled to obtain synchronized point cloud data for multiple targets; and

[0054] During the data processing stage, static point cloud data and synchronous point cloud data with the same target information are analyzed separately to determine multiple sets of dynamic performance parameters of the lidar for multiple targets.

[0055] According to embodiments of this disclosure, the lidar performance testing method further includes:

[0056] A new preset speed is determined by adjusting the turntable rotation speed and / or the distance between the target and the turntable's rotation axis, and the dynamic performance of the lidar is tested at this new preset speed; and

[0057] Dynamic performance testing of all lidar field-of-view zones was achieved by adjusting the installation angle of the lidar mounted on the fixed lidar module.

[0058] In another aspect, this disclosure also provides a lidar performance testing device, such as... Figure 2 As shown, the testing apparatus includes:

[0059] The radar fixing module includes a bracket and an adjustment platform. The bracket is configured to be fixed relative to the turntable position. The adjustment platform is configured to be fixedly connected to the bracket via a first end and to the lidar via a second end. By adjusting the tilt angle of the adjustment platform, the detection position is positioned at the center of the lidar's field of view partition. The lidar's field of view partition is one of multiple lidar field of view partitions determined by dividing the lidar's field of view. The distance between the bracket and the turntable can be adjusted according to the actual test conditions, for example, it can be 10m, 20m, 30m, 40m, 100m or more.

[0060] A rotating disc mechanism includes a turntable and a drive mechanism. The turntable is configured to drive a target positioned on its edge in a circular motion under the drive of the drive mechanism. The target's motion path is determined by a detected position. Figure 2 As shown, the radar's emission direction is directly facing the turntable, and the detection position is set directly above the turntable. It should be noted that the detection position is not limited to this and can be set according to the actual test conditions.

[0061] The timing synchronization module is configured to detect the motion state of the target to determine the motion timing data of the target, and to determine the synchronization timing signal based on the motion timing data of the target and the position information of the detection position.

[0062] The control module is connected to the disk rotation mechanism, the lidar, and the timing synchronization module. It is configured to acquire static point cloud data and synchronized point cloud data, and to determine the dynamic performance parameters of the lidar field of view partition based on the static point cloud data and synchronized point cloud data.

[0063] According to an embodiment of this disclosure, the edge of the turntable is provided with at least one mounting hole for mounting at least one target;

[0064] According to an embodiment of this disclosure, the control module includes a speed control unit, a static data acquisition unit, a synchronization timing determination unit, a synchronization point cloud data acquisition unit, and a performance analysis unit.

[0065] The speed control unit is configured to control the rotational speed of the turntable based on a communication connection with the turntable rotation mechanism; the static data acquisition unit obtains static point cloud data based on a communication connection with the lidar; the synchronization timing determination unit determines the synchronization timing signal based on a communication connection with the timing synchronization module; the synchronization point cloud data acquisition unit is configured to control the lidar to perform timing synchronization data acquisition in response to the synchronization timing signal, so as to determine the synchronization point cloud data; and the performance analysis unit is configured to determine the dynamic performance parameters of the lidar field of view partition based on the static point cloud data and the synchronization point cloud data.

[0066] According to an embodiment of this disclosure, the timing synchronization module includes a rotation angle measurement unit and a synchronization signal generation unit; the rotation angle measurement unit is, for example, disposed on the rotating shaft of a turntable and configured to collect and output real-time rotation angle data of the turntable; the synchronization signal generation unit is configured to receive the real-time rotation angle data and determine a synchronization timing signal based on the real-time rotation angle data and the detection position.

[0067] According to an embodiment of this disclosure, an electric motor drives a turntable to rotate via a gear set. Several reflective targets are mounted on the turntable. A light-shielding sticker can be attached to the turntable; as the turntable rotates, the sticker generates a periodic signal in front of a photoelectric sensor, thus calculating the angular velocity. A lidar is mounted on a bracket and scans the turntable plane. When a target passes through the radar scanning area, the radar generates a dynamic echo signal. The control module synchronizes the radar sampling timing with the turntable rotation signal to calculate target detection time delay, distance and angle errors, point cloud trajectory stability, etc. During testing, targets with different linear velocities can be simulated by changing the electric motor speed. For example, when the turntable radius is 0.5m and the speed is 600rpm, the target linear velocity is approximately 31.4m / s, which can simulate a high-speed vehicle.

[0068] This device can be used in a laboratory environment to quickly evaluate the dynamic performance of vehicle-mounted lidar at different speeds, providing basic data for algorithm calibration and hardware optimization.

[0069] Existing motion distortion evaluation methods mostly focus on: frame consistency of point clouds, which indirectly reflects distortion through ICP registration error or trajectory smoothness; and scan path error, which is a macroscopic assessment of reconstruction errors in fixed scenes. However, the "velocity-related centerline segmented deformation index" proposed in this invention is directly based on the spatial offset and segmented angle difference of the target's geometric center axis. It can independently reveal the radar's time synchronization error, scanning timing delay, and ranging-time decoupling characteristics under high-speed conditions, and is a quantitative parameter with clear physical meaning and engineering measurability. Compared with traditional indicators, the technical solution disclosed in this invention has the advantages of quantifying dynamic distortion for a single target; it can directly correspond to radar design parameters (scanning cycle, number of lines, sampling delay, etc.); and it is applicable to general testing of various structural optical systems (mechanical, MEMS, solid-state scanning radar). The technical solution disclosed in this invention provides a standardized measurement method for dynamic performance evaluation, production consistency detection, and calibration algorithm verification of vehicle-mounted LiDAR, supporting the quantitative verification of motion compensation algorithms under different speed scenarios; and guiding radar structure optimization and timing control system design, especially in the fields of high-speed perception and intelligent driving, improving the equipment's ability to accurately model dynamic scenes. Therefore, this technology has important engineering practical value and standardization significance in the dynamic performance measurement of lidar, and fills a key gap in the existing distortion index system.

[0070] The disclosed lidar performance testing method and apparatus employs an adjustable-speed turntable mechanism, capable of simulating moving targets at different speeds. The speed range covers typical traffic scenarios (e.g., equivalent relative speeds of 5 m / s–40 m / s), overcoming the limitations of traditional static testing. To simulate target motion at different vehicle speeds, the rotating turntable radius R and rotational speed n of the apparatus can be designed based on the target linear velocity formula v = 2πRn. The turntable radius is adjustable from 0.2 m to 0.8 m, and the motor speed is designed to range from 30 rpm to 900 rpm. Through combined adjustments, a dynamic range of target linear velocities from 0.5 m / s to 125 m / s can be covered, meeting the dynamic response testing requirements of lidar in passenger vehicles and high-speed scenarios. For example, when the radius is 0.5 m and the rotational speed is 600 rpm, the target linear velocity is approximately 31.4 m / s, which can be used to simulate a target reflection scenario moving at 110 km / h. This design range ensures the mechanical safety of the apparatus while covering the upper limit of detection speed for typical vehicle-mounted radars, providing a reference for performance curve calibration. The turntable allows for quick replacement of test targets with different reflectivities, shapes, or materials to simulate various objects (such as vehicles, roadblocks, and pedestrians). To accommodate test targets with different reflective characteristics and shapes, this invention features a modular target mounting interface. Multiple standardized mounting holes (equally distributed) are pre-drilled on the outer edge of the turntable, allowing for the rapid installation of different types of test targets, such as triangular pyramidal reflectors, diffuse reflective plates, or cylindrical rod-shaped targets. Each target module is secured by a quick-release slot and positioning pin, with a replacement time of no more than 30 seconds. The back of the target is equipped with an identification code or reflective marking for automatically recording target type information, thus establishing a correspondence between test data and target parameters. This structure facilitates repeated verification in multiple scenarios, significantly improving the flexibility and scalability of the testing system. Time synchronization with the lidar sampling frames is achieved through photoelectric shielding stickers or rotary encoders, enabling quantitative evaluation of point cloud update delay, distance offset, and dynamic resolution. To ensure synchronization between the lidar sampling frames and the target's movement position, this invention includes a synchronization trigger and timing calibration module. This module includes a photoelectric encoder or Hall sensor mounted on the rotating shaft for real-time output of angular position signals. The control unit acquires angle signals via a high-speed counting interface (≥10 kHz) and synchronizes them with the lidar acquisition system. Synchronization can be achieved through hardware triggering (TTL level signal) or software timestamp comparison. The overall system timing accuracy is better than ±0.5ms, effectively suppressing phase errors caused by differences in radar frame rates. By matching the target angular position with the radar detection results, the range drift and response delay of the radar under dynamic targets can be accurately evaluated. To ensure measurement accuracy, this invention proposes a standardized calibration and verification process. During the calibration phase, the radius of the target's circular trajectory is measured using a laser rangefinder or a high-precision optical positioning system, and the correspondence between the encoder output and the actual angle is calibrated. Subsequently, radar reference sampling is performed at low speed to calculate the trajectory consistency and distance error of the target detection point.The verification phase employs repeatable tests at multiple rates (low, medium, and high). By comparing point cloud data, the dynamic accuracy and stability of the system are verified under dynamic conditions through point cloud deformation. This process ensures the traceability and performance consistency of the testing equipment during long-term use.

[0071] The embodiments of this disclosure have been described in detail above with reference to the accompanying drawings. It should be noted that implementations not illustrated or described in the drawings or the main text of the specification are forms known to those skilled in the art and have not been described in detail. It should be understood that the above are merely specific embodiments of this disclosure and are not intended to limit this disclosure. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this disclosure should be included within the protection scope of this disclosure.

Claims

1. A method for testing the performance of a lidar system, characterized in that, include: Set the detection position and adjust the installation angle of the lidar mounted on the lidar fixed module so that the detection position is at the center of the lidar field of view partition. The lidar field of view partition is one of multiple lidar field of view partitions determined by dividing the lidar field of view. The lidar is used to detect a statically positioned target at the detection location to determine static point cloud data; The drive turntable rotates at a preset angular velocity, thereby causing the target set on the edge of the turntable to perform circular motion at a preset speed, wherein the motion path of the target passes through the detection position; The motion state of the target is detected by a timing synchronization module to determine the motion timing data of the target; A synchronization timing signal is determined based on the motion timing data of the target and the position information of the detection location; In response to the synchronization timing signal, synchronized point cloud data is obtained by timing synchronization data acquisition using lidar. Based on the static point cloud data and the synchronous point cloud data, the dynamic performance parameters of the lidar field of view partition are determined, and the lidar performance test is completed.

2. The performance testing method according to claim 1, characterized in that, The determination of the dynamic performance parameters of the lidar field of view partition based on the static point cloud data and the synchronous point cloud data includes: The static point cloud data is corrected to obtain corrected static point cloud data; A fitting operation is performed based on the corrected static point cloud data to determine the reference center axis of the target; Based on the synchronized point cloud data, N synchronized point cloud data groups are obtained by grouping the data, where N≥2; The N synchronous point cloud data groups are each fitted to determine N segmented center axes; Based on the reference center axis and the N segmented center axes, the dynamic performance parameters of the lidar field of view partition are determined.

3. The performance testing method according to claim 1, characterized in that, The target's motion time-series data includes the target's real-time position data. The step of determining a synchronization time-series signal based on the target's motion time-series data and the position information of the detection location includes: The motion timing data is detected, and when the real-time position data of the target is the same as the position information of the detected position, the synchronization timing signal is generated.

4. The performance testing method according to claim 1, characterized in that, The process of acquiring synchronized point cloud data via lidar in response to the synchronization timing signal includes: In response to the synchronization timing signal, the control module controls the lidar to acquire point cloud data to obtain intermediate point cloud frame data. The intermediate point cloud frame data is extracted to determine the target point cloud cluster, thereby obtaining synchronized point cloud data.

5. The performance testing method according to claim 2, characterized in that, The dynamic performance parameters of the lidar field of view partition include: tilt angle, distortion, and dynamic distortion composite index; the method for determining the dynamic performance parameters of the lidar field of view partition includes: The tilt angle is determined based on the reference center axis and the N segmented center axes; Based on the N segmented center axes, determine the N-1 segment angles between adjacent segmented center axes; The degree of twist is determined based on the N-1 segmented angles; The dynamic distortion comprehensive index is determined based on the tilt angle and the twist degree.

6. The performance testing method according to claim 1, characterized in that, When there are multiple targets, the method further includes: When acquiring static point cloud data, each target is detected to obtain multiple static point cloud data, and the multiple static point cloud data are marked with corresponding target information. When acquiring the synchronization timing signal, the target recognition module identifies each target passing through the detection position to obtain the target information, and uses the target information to mark the corresponding synchronization timing signal to obtain the multi-target synchronization timing signal. When determining the synchronized point cloud data, synchronized point cloud data is collected in response to the multi-target synchronization timing signal, and the synchronized point cloud data is marked based on the target information corresponding to the synchronization timing signal to obtain synchronized point cloud data of multiple targets. During the data processing stage, the static point cloud data and the synchronous point cloud data with the same target information are analyzed to determine multiple sets of dynamic performance parameters of the lidar for multiple targets.

7. The performance testing method according to claim 1 or 6, characterized in that, Also includes: A new preset speed is determined by adjusting the rotation speed of the turntable and / or the distance between the target and the rotation axis of the turntable, and the dynamic performance of the lidar is tested at the new preset speed. Dynamic performance testing of all lidar field-of-view zones was achieved by adjusting the installation angle of the lidar mounted on the fixed lidar module.

8. A lidar performance testing device, characterized in that, include: The radar fixing module includes a bracket and an adjustment platform. The bracket is configured to be fixed relative to the position of the turntable. The adjustment platform is configured to be fixedly connected to the bracket via a first end and fixedly connected to the lidar via a second end. By adjusting the tilt angle of the adjustment platform itself, the detection position is positioned at the center of the lidar field of view partition. The lidar field of view partition is one of multiple lidar field of view partitions determined by dividing the lidar field of view. A disc rotation mechanism includes a turntable and a drive mechanism. The turntable is configured to drive a target disposed on the edge of the turntable to perform a circular motion under the drive of the drive mechanism, wherein the motion path of the target passes through the detection position. The timing synchronization module is configured to detect the motion state of the target to determine the motion timing data of the target, and to determine a synchronization timing signal based on the motion timing data of the target and the position information of the detection position. The control module is communicatively connected to the disk rotation mechanism, the lidar, and the timing synchronization module, and is configured to acquire static point cloud data and synchronized point cloud data, and determine the dynamic performance parameters of the lidar field of view partition based on the static point cloud data and synchronized point cloud data.

9. The performance testing apparatus according to claim 8, characterized in that, The edge of the turntable is provided with at least one mounting hole for mounting at least one of the targets; The control module includes: Speed ​​control unit: controls the rotational speed of the turntable based on a communication connection with the turntable rotation mechanism; Static data acquisition unit: Based on the communication connection with the lidar, it acquires static point cloud data; Synchronization timing determination unit: Based on the communication connection with the timing synchronization module, it determines the synchronization timing signal; Synchronous point cloud data acquisition unit: In response to the synchronous timing signal, controls the lidar to perform timing synchronization data acquisition in order to determine the synchronous point cloud data; Performance analysis unit: Determines the dynamic performance parameters of the lidar field of view partition based on the static point cloud data and the synchronous point cloud data.

10. The performance testing apparatus according to claim 8, characterized in that, The timing synchronization module includes a rotation angle measurement unit and a synchronization signal generation unit disposed on the rotating shaft of the turntable; The rotation angle measurement unit is configured to collect and output real-time rotation angle data of the turntable; The synchronization signal generation unit is configured to receive the real-time rotation angle data and determine a synchronization timing signal based on the real-time rotation angle data and the detection position.

Images