Laser radar test method based on long-distance target distance measurement result

Abstract

The invention discloses a laser radar test method based on a long-distance target distance measurement result, and the method comprises the following steps: obtaining a laser radar echo signal, and extracting the frequency spectrum information of the echo signal; doppler frequency shift characteristic parameters are extracted according to the frequency spectrum information, and the relative speed of the target is calculated based on a preset frequency shift calculation model; acquiring multi-source sensor data associated with the laser radar test platform, and fusing to construct a dynamic ranging reference coordinate system; performing dynamic motion compensation on a laser ranging result according to the relative speed of the target to obtain a target initial measurement distance after Doppler correction; mapping the target initial measurement distance after Doppler correction to the dynamic distance measurement reference coordinate system to obtain a remote target final distance measurement result aligned with a test platform coordinate system; and outputting the accurate position and moving state information of the target in the three-dimensional space according to the final distance measurement result. According to the invention, the ranging precision and stability can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of lidar testing, and particularly to a lidar testing method based on ranging results of a long-distance target. Background Art

[0002] As a high-precision, non-contact ranging technology, lidar has been widely used in recent years in fields such as autonomous driving, surveying and mapping, security monitoring, and intelligent transportation. Its basic principle is to emit a laser beam and receive the target echo, and calculate the target distance according to the time that the laser travels in the medium. However, in the long-distance ranging scenario, the intensity of the laser signal attenuates significantly, the target echo signal is vulnerable to environmental interference, and at the same time, the relative movement of the target will introduce additional error factors, resulting in insufficient stability and accuracy of the ranging result.

[0003] Most of the existing lidar testing methods are based on static coordinate systems for measurement, ignoring the influence of the dynamic movement of the testing platform itself on the ranging result. Especially in dynamic testing environments such as vehicles and drones, the relative movement state between the laser and the target may cause a significant amplification of the ranging error. In addition, when the target is in a high-speed movement state, due to the Doppler frequency shift effect of the laser reflection wave, the reflection signal is distorted in the time domain and frequency domain, which will further affect the ranging accuracy.

[0004] At the same time, traditional ranging methods often rely on the information of a single sensor, lack the fusion of multi-source data, and cannot construct a complete and dynamically changing reference coordinate system in the real world, resulting in the testing accuracy of lidar in complex or variable environments not meeting the requirements of high-precision applications. For example, in urban roads, complex obstacle environments, or scenarios where the target moves violently, the ranging results often show fluctuations and jumps, seriously affecting the calibration, performance verification, and stability analysis of the lidar system. Summary of the Invention

[0005] The purpose of this part is to outline some aspects of the embodiments of the present invention and briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this part, as well as in the abstract and title of the present application, to avoid obscuring the purpose of this part, the abstract, and the title. However, such simplifications or omissions cannot be used to limit the scope of the present invention.

[0006] To solve the above technical problems, the present invention provides the following technical solution: a lidar testing method based on ranging results of a long-distance target, including the following steps: Obtain the lidar echo signal and extract the spectral information of the echo signal; Extract the Doppler frequency shift characteristic parameters according to the spectral information and calculate the relative velocity of the target based on a preset frequency shift resolution model ; Obtain multi-source sensor data associated with the lidar test platform and fuse it to construct a dynamic ranging reference coordinate system; Perform dynamic motion compensation on the lidar ranging result according to the relative speed of the target to obtain the initial measured distance of the target after Doppler correction; Map the initial measured distance of the target after Doppler correction to the dynamic ranging reference coordinate system to obtain the final ranging result of the long-distance target aligned with the test platform coordinate system; Output the accurate position and movement status information of the target in three-dimensional space according to the final ranging result.

[0007] As a preferred solution of the lidar test method based on the long-distance target ranging result of the present invention, wherein: the frequency shift resolution model is based on the laser wavelength and the frequency shift value , and the relative speed is calculated using the following relational expression: , wherein, is the relative speed of the target, and the obtained value is used for dynamic ranging compensation, is the Doppler frequency shift amount, c is the speed of light, is the emission frequency of the laser.

[0008] As a preferred solution of the lidar test method based on the long-distance target ranging result of the present invention, wherein: the multi-source sensor data includes but is not limited to IMU data, GPS data, and visual sensor image data.

[0009] As a preferred solution of the lidar test method based on the long-distance target ranging result of the present invention, wherein: obtaining the multi-source sensor data associated with the lidar test platform and fusing it to construct a dynamic ranging reference coordinate system includes the following steps: Calculate the attitude change information of the test platform according to the IMU data; Obtain the position information of the test platform according to the GPS data; Construct a visual SLAM image coordinate system of the environment where the platform is located according to the visual sensor image data; Based on the fusion of the IMU data, the GPS data, and the visual SLAM image coordinate system data, construct a dynamic ranging reference coordinate system.

[0010] As a preferred solution of the lidar test method based on the long-distance target ranging result of the present invention, wherein: based on the fusion of the IMU data, the GPS data, and the visual SLAM image coordinate system data, constructing a dynamic ranging reference coordinate system includes the following steps: Taking the visual SLAM image coordinate system as the initial environmental reference system, using the attitude change amount deduced by the IMU as the intermediate state constraint, and using the position information of the test platform provided by the GPS as the global position factor, a fusion optimization model is constructed, and the fusion optimization model is a factor graph model; Using a non-linear optimization algorithm to globally optimize the above factor graph, obtaining the optimal state estimation results of the platform at each moment, including three-dimensional position coordinates (X, Y, Z) and three-axis attitude angles; Taking the platform state at the current moment output by the factor graph optimization as the coordinate system benchmark, a three-dimensional dynamic ranging reference coordinate system is constructed.

[0011] As a preferred solution of the lidar test method based on the long-distance target ranging result of the present invention, wherein: dynamically compensating the laser ranging result according to the relative speed of the target to obtain the initial measured distance of the target after Doppler correction, including the following steps: Obtain the original lidar ranging result and determine the data format and unit of the original ranging result; According to the calculated relative speed of the target determine the time correction amount required for compensation ; Based on the compensation time correction amount combining with the relative speed of the target calculate the ranging offset distance caused by the movement of the target ; Subtract the ranging offset distance from the original ranging result to obtain the corrected distance after Doppler motion compensation , and the specific calculation formula is: ; Perform filtering and smoothing processing on the corrected distance after Doppler motion compensation and output the initial measured distance of the target after Doppler correction.

[0012] As a preferred solution of the lidar test method based on the long-distance target ranging result of the present invention, wherein: the step of determining the time correction amount required for compensation according to the calculated relative speed of the target includes the following steps: Calculate the one-way propagation time of the laser signal according to the speed of light constant c , and this propagation time is used to estimate the actual displacement of the target during the laser propagation period; Multiply the relative speed of the target by the one-way propagation time to obtain the time correction amount required for compensation Determine the displacement of the target during signal propagation by association , and the formula is: , and this displacement reflects the spatial offset of the target due to its own movement during the round-trip of the laser; Convert the displacement into the corresponding time correction amount .

[0013] As a preferred solution of the lidar test method based on the long-distance target ranging result described in the present invention, wherein: map the initially measured distance of the target after Doppler correction to the dynamic ranging reference coordinate system to obtain the final ranging result of the long-distance target aligned with the test platform coordinate system, including the following steps: Obtain the initially measured distance of the target after Doppler correction data, and determine the data format and coordinate reference system attributes of the initially measured distance of the target; Obtain the dynamic ranging reference coordinate system information output by factor graph optimization at the current moment, including the three-dimensional position and attitude angle of the platform; According to the installation pose parameters of the lidar on the test platform, construct the pose transformation matrix between the local coordinate system to which the original ranging data belongs and the platform coordinate system; Convert the initially measured distance data of the target into the three-dimensional position coordinates of the target in the local coordinate system of the test platform through the coordinate transformation matrix; According to the current global state information of the test platform, map the three-dimensional position coordinates of the target in the local coordinate system of the platform to the dynamic ranging reference coordinate system; Output the three-dimensional coordinate values of the target with spatial alignment completed as the final ranging result of the long-distance target aligned with the test platform coordinate system.

[0014] As a preferred solution of the lidar test method based on the long-distance target ranging result described in the present invention, wherein: output the accurate position and movement state information of the target in three-dimensional space according to the final ranging result, including the following steps: Obtain the three-dimensional coordinate values of the target that have been mapped to the end of the dynamic ranging reference coordinate system ; Combine the ranging results of multiple consecutive moments to construct a target space trajectory sequence; Based on the target space trajectory sequence, calculate the dynamic motion state information of the target and output the real-time position and motion state of the target in three-dimensional space; The dynamic motion state information includes velocity vector, acceleration vector and motion direction.

[0015] As a preferred solution of the lidar test method based on the ranging results of long-distance targets in the present invention, wherein: the calculation of the dynamic motion state information is based on the ranging results of the target at multiple consecutive moments, and the velocity change trend and acceleration characteristics of the target are extracted through the position difference method, Kalman filter or other state estimation algorithms, so as to reflect the motion behavior of the target in real time.

[0016] Advantages of the present invention: 1. By introducing the Doppler frequency shift feature extraction and frequency shift resolution model, the present invention calculates the relative velocity of the target and performs dynamic motion compensation on the lidar ranging results, enabling the ranging process to adapt to the ranging errors caused by the high-speed movement of the target. Compared with the traditional ranging methods based on static or uniform speed models, the present invention can effectively solve the ranging offset problem caused by high-speed targets, and significantly improve the ranging accuracy and stability in long-distance dynamic scenarios; 2. The present invention fuses the data of IMU, GPS and visual SLAM image coordinate systems to construct a dynamic ranging reference coordinate system, enabling the lidar ranging results to be aligned with the spatial state changes of the test platform in real time. Compared with the traditional methods relying on fixed reference systems, the present invention has higher environmental adaptability and robustness, and can effectively cope with the coordinate system disturbance problems during complex dynamic processes such as the movement, turning and vibration of the test platform, ensuring the spatial consistency and time synchronization of the ranging data; 3. The present invention not only outputs the position coordinates of long-distance targets in three-dimensional space, but also further calculates the dynamic information such as the velocity and acceleration of the target based on the ranging sequences at multiple moments, enabling the functions of target behavior analysis and trajectory tracking. This extended ability endows the present invention with the potential to expand to higher-order application scenarios such as target recognition, motion prediction, and dynamic obstacle detection, significantly improving the practical value of the lidar system in intelligent transportation, dynamic monitoring and environmental perception. Description of the drawings

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention, and those of ordinary skill in the art can obtain other drawings without creative efforts. Among them: Figure 1 It is the overall flow block diagram of the lidar test method based on the ranging results of long-distance targets in the present invention. Detailed implementation manners

[0018] In order to make the above objects, features and advantages of the present invention more obvious and understandable, the detailed implementation manners of the present invention will be described in detail below with reference to the drawings in the specification.

[0019] In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention. However, the present invention may be practiced in other ways than those specifically described herein, and those skilled in the art may make similar extensions without departing from the spirit of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed below.

[0020] Secondly, as used herein, "an embodiment" or "embodiments" refer to specific features, structures, or characteristics that may be included in at least one implementation of the present invention. The appearances of "in an embodiment" in different places in this specification do not all refer to the same embodiment, nor are they separate or alternative embodiments that exclude each other.

[0021] Thirdly, the present invention is described in detail in conjunction with schematic diagrams. When detailing the embodiments of the present invention, for ease of illustration, the cross-sectional views showing the device structure are enlarged locally in a non-generalized scale, and the schematic diagrams are only examples and should not limit the scope of protection of the present invention herein. In addition, in actual production, three-dimensional spatial dimensions including length, width, and depth should be included.

[0022] Embodiment 1 Referring to Figure 1 , an embodiment of the present invention provides a lidar test method based on the ranging result of a long-distance target, including the following steps: S1: Obtain the lidar echo signal and extract the spectral information of the echo signal.

[0023] Specifically, during the spectral information extraction process, preprocess the echo signal, such as denoising filtering, window function weighting, etc., to improve the accuracy and stability of spectral analysis. In this way, the spectral features are efficiently extracted from the lidar echo, laying a foundation for the subsequent dynamic compensation and precise ranging of this application.

[0024] It should be noted that the echo signal is received by the receiving system of the lidar to form the original echo signal data. This echo signal contains information such as the distance, speed, and reflection intensity of the target object, which is the basis for subsequent distance calculation and speed estimation.

[0025] In order to further analyze the motion characteristics of the target and improve the ranging accuracy, spectral extraction processing is performed on the original echo signal. Specifically, frequency domain analysis methods such as fast Fourier transform are used to process the echo signal, converting it from a time-domain signal to a frequency-domain signal, thereby obtaining spectral information containing Doppler frequency shift components. This spectral information can clearly characterize the frequency change characteristics contained in the echo signal, especially the Doppler frequency shift components caused by the target motion. By analyzing features such as the main peak position and energy distribution in the spectrum, it provides a basic support for subsequent Doppler frequency shift extraction and target relative speed calculation.

[0026] S2: Extract the Doppler frequency shift characteristic parameters based on the spectrum information, and calculate the relative velocity of the target based on a preset frequency shift calculation model. .

[0027] It should be noted that the frequency shift characteristic parameters are extracted through spectrum analysis. The Doppler frequency shift characteristic parameters include the main frequency offset, which is the Doppler frequency shift generated by the laser echo due to the relative movement of the target. The Doppler frequency shift characteristic parameters reflect the radial velocity of the target relative to the lidar system and are the key to achieving high-precision dynamic ranging and motion compensation.

[0028] Specifically, the frequency shift calculation model is based on the laser wavelength and the frequency shift value , and the relative velocity is calculated using the following relational expression: , where is the relative velocity of the target, and the obtained value is used for dynamic ranging compensation. is the Doppler frequency shift amount, c is the speed of light, is the emission frequency of the laser.

[0029] Through the above method, the present invention can accurately extract the Doppler characteristics of the target from the spectrum information, and calculate the real-time motion speed of the target relative to the lidar system through the frequency shift calculation model, providing key data support for subsequent dynamic compensation and distance mapping.

[0030] S3: Obtain multi-source sensor data associated with the lidar test platform, and fuse and construct a dynamic ranging reference coordinate system to provide a unified spatial alignment basis for the lidar ranging results.

[0031] The multi-source sensor data includes but is not limited to inertial measurement unit (IMU) data, global positioning system (GPS) data, and visual sensor image data.

[0032] Specifically, obtaining the multi-source sensor data associated with the lidar test platform and fusing and constructing a dynamic ranging reference coordinate system includes the following steps: S31: Calculate the attitude change information of the test platform based on the IMU data.

[0033] It should be noted that the inertial measurement unit (IMU) includes but is not limited to gyroscopes and accelerometers, which are used to measure the platform's three-axis angular velocity, acceleration and other motion state information in real time. By integrating and filtering the IMU data, combined with algorithms such as Kalman filtering or extended Kalman filtering, the platform's attitude change information during the ranging process can be obtained, mainly including pitch angle (pitch), roll angle (roll) and yaw angle (yaw). This attitude change information serves as an important basis for subsequent coordinate transformation and error compensation, and helps to improve the accuracy of laser ranging space alignment.

[0034] S32: Obtain location information of the test platform according to GPS data.

[0035] It should be noted that the Global Positioning System (GPS) can provide the platform's location information in a geographic coordinate system, and the typical data form is three-dimensional coordinates (longitude, latitude, and elevation).

[0036] Preferably, differential GPS or real-time kinematic positioning (RTK) technology is used to obtain position information with high accuracy. This high-precision position information provides a global position reference for constructing a range-finding reference coordinate system and is used to constrain position variables in the factor graph.

[0037] S33: Construct a visual SLAM image coordinate system of the environment where the platform is located according to the visual sensor image data.

[0038] It should be noted that the image data collected by the visual sensor (such as a monocular or binocular camera) can be used to construct a spatial model of the environment where the platform is located. Preferably, through the simultaneous localization and mapping (SLAM) technology, through feature point matching, depth estimation and geometric recovery between image frames, a sparse or dense environmental point cloud map can be obtained, and then a visual SLAM image coordinate system can be constructed. This coordinate system, as an environmental reference system, can be used with IMU and GPS data to constrain and compensate for drift errors, thereby improving the robustness and stability of the overall system.

[0039] S34: Based on the fusion of IMU data, GPS data and visual SLAM image coordinate system data, a dynamic ranging reference coordinate system is constructed. Specifically, the following steps are included: S34-1: Use the visual SLAM image coordinate system as the initial environment reference system, the attitude change derived by the IMU as the intermediate state constraint, and the position information of the test platform provided by the GPS as the global position factor to build a fusion optimization model.

[0040] Preferably, the fusion optimization model is a factor graph model, whose nodes represent the state (position + posture) of the platform at different times, and whose edges represent the constraint factors introduced by each sensor data. Optionally, a graph neural network structure is introduced to dynamically learn and optimize the weights between factors to improve the fusion accuracy and robustness.

[0041] S34-2: Use a non-linear optimization algorithm (such as g2o, Ceres Solver, etc.) to globally optimize the above factor graph to obtain the optimal state estimation results of the platform at each moment, including three-dimensional position coordinates (X, Y, Z) and three-axis attitude angles (roll, pitch, yaw). This optimization process can effectively fuse multi-source data and offset their respective random errors and systematic errors.

[0042] S34-3: Take the platform state at the current moment output by the factor graph optimization as the coordinate system reference to construct a three-dimensional dynamic ranging reference coordinate system. Define the coordinate origin as the center of mass of the platform, the X-axis towards the forward direction of the platform, the Y-axis pointing to the left side of the platform, and the Z-axis perpendicular upward. This coordinate system can be updated in real time as the platform moves and provides a unified spatial alignment reference for subsequent lidar ranging data.

[0043] S4: Perform dynamic motion compensation on the lidar ranging results according to the relative velocity of the target to obtain the initial measured distance of the target after Doppler correction.

[0044] Specifically, performing dynamic motion compensation on the lidar ranging results according to the relative velocity of the target to obtain the initial measured distance of the target after Doppler correction includes the following steps: S41: Obtain the original lidar ranging results , and determine the data format and unit of the original ranging results; ensure correct data parsing, unified units, compatible compensation logic, and multi-source fusion compatibility, thereby laying a foundation for subsequent dynamic motion compensation. In addition, the original lidar ranging results As the uncorrected initial distance value, it is the basis for calculating the Doppler compensation amount of the target.

[0045] S42: Determine the time correction amount required for compensation according to the relative velocity of the target calculated in step S2 ; Specifically, according to the relative velocity of the target calculated in step S2 , determine the time correction amount required for compensation , including the following steps: S42-1: Calculate the one-way propagation time of the laser signal according to the speed of light constant c , and this propagation time is used to estimate the actual displacement of the target during the laser propagation period; S42-2: Relate the relative velocity of the target to the one-way propagation time to determine the displacement of the target during the signal propagation period , and the formula is: , and this displacement Reflect the spatial offset generated by the target's own movement during the round-trip of the laser; S42-3: Convert the displacement into the corresponding time correction amount , preferably, determined by the formula ; optionally, if the relative velocity of the target varies dynamically with time, then use for correction calculation, where "+" is used for the target moving away from the radar and "-" is used for the target approaching the radar; the output is used as the dynamic compensation time parameter to correct the ranging offset of the target.

[0046] S43: According to the compensation time correction amount , combined with the relative velocity of the target, calculate the ranging offset distance caused by the target movement, and the formula is: ; the ranging offset distance represents the laser signal propagation path error caused by the target movement and is a key parameter for subsequent correction.

[0047] S44: Perform a difference operation on the ranging offset distance and the original ranging result to obtain the corrected distance after Doppler motion compensation, and the specific calculation formula is: . The core of this step is to back-calculate the position where the target should be between laser emission and reception, thereby significantly improving the ranging accuracy in a dynamic target scenario.

[0048] It should be noted that the corrected distance after Doppler motion compensation corresponds one-to-one with the dynamic ranging reference coordinate system constructed in step S3 and is synchronized through timestamps to ensure accurate alignment of coordinate mapping (S5).

[0049] S45: Perform filtering and smoothing processing on the corrected distance after Doppler motion compensation, and output the preliminary measured distance of the target after Doppler correction; It should be noted that filtering and smoothing processing are used to reduce accidental ranging fluctuations caused by relative velocity estimation errors or system noise.

[0050] Preferably, algorithms such as moving average, Gaussian filtering, or Kalman filtering can be used to achieve smooth output of distance data. After filtering, further unified formatting processing can be performed on to ensure data structure compatibility and consistency in subsequent coordinate system mapping steps.

[0051] S5: Map the Doppler-corrected initial measured distance of the target into the dynamic ranging reference coordinate system to obtain the final ranging result of the long-distance target aligned with the coordinate system of the test platform.

[0052] Specifically, mapping the Doppler-corrected initial measured distance of the target into the dynamic ranging reference coordinate system to obtain the final ranging result of the long-distance target aligned with the coordinate system of the test platform includes the following steps: S51: Obtain the Doppler-corrected initial measured distance data of the target, and determine the data format and coordinate reference system attributes of the initial measured distance of the target; In this step, the initial measured distance data obtained by Doppler effect correction is used to reflect the relative distance change of the target. After this data is Doppler-corrected, it will reflect the actual distance of the target relative to the measurement platform. To ensure the accuracy of data conversion, it is necessary to first determine the data format and coordinate reference system attributes of the initial measured distance of the target. This includes analyzing the type, precision of the data, and its definition in the space coordinate system to ensure that the data can be compatible and convertible with the coordinate system of the test platform.

[0053] S52: Obtain the dynamic ranging reference coordinate system information output by factor graph optimization at the current moment, including the three-dimensional position and attitude angles of the platform.

[0054] In this step, obtaining the dynamic ranging reference coordinate system information at the current moment is crucial, especially the three-dimensional position and attitude angles of the platform. The information output by the factor graph optimization algorithm will provide the precise position information (i.e., the three-dimensional coordinates of the platform) and attitude angles (such as the pitch, yaw, and roll angles of the platform) of the test platform. These information are the basis for target position transformation and mapping because in the subsequent steps, the initial ranging of the target needs to be mapped into the coordinate system of the test platform.

[0055] S53: According to the installation pose parameters of the lidar on the test platform, construct the pose transformation matrix between the local coordinate system to which the original ranging data belongs and the platform coordinate system; In this step, it is necessary to construct the coordinate transformation matrix according to the installation pose of the lidar on the test platform. The installation pose parameters of the lidar include the translation amount and rotation matrix of the lidar relative to the test platform coordinate system. The translation amount represents the position of the lidar installation point in the platform coordinate system, while the rotation matrix represents the rotation state of the lidar relative to the platform coordinate system. These parameters will help to transform from the local coordinate system of the lidar to the test platform coordinate system to ensure the spatial alignment and accurate conversion of the data.

[0056] S54: Convert the initial measured distance data of the target into the three-dimensional position coordinates of the target in the local coordinate system of the test platform through the coordinate transformation matrix.

[0057] In this step, through the constructed pose transformation matrix, the initially measured distance of the target after Doppler correction is transformed from the local coordinate system of the lidar to the local coordinate system of the test platform. This transformation is necessary because the ranging data of the lidar is initially collected relative to the position of the lidar itself. Through this step, the three-dimensional position coordinates of the target can be transformed to a local coordinate system related to the test platform, providing a basis for subsequent further calculations and coordinate mapping.

[0058] S55: According to the current global state information of the test platform, map the three-dimensional position coordinates of the target in the local coordinate system of the platform to the dynamic ranging reference coordinate system.

[0059] It should be noted that the global state information includes factors such as the dynamic position and attitude angle of the platform, ensuring that the position of the target not only reflects the relative position of the target with respect to the platform but also takes into account the dynamic changes of the platform. Through this step, the target can be accurately located in the changing coordinate reference system, ensuring the timeliness and accuracy of the ranging results.

[0060] S56: Output the three-dimensional coordinate values of the target with completed spatial alignment as the final ranging result of the long-distance target aligned with the coordinate system of the test platform, which can be directly used for subsequent data analysis, target tracking, and other applications.

[0061] In step S5, accurate ranging of long-distance targets can be ensured, the ranging error caused by the Doppler effect is eliminated, and the consistency of the coordinate system of the test platform is guaranteed through dynamic coordinate transformation, providing reliable data support for subsequent target tracking and positioning.

[0062] S6: Output the accurate position and movement state information of the target in three-dimensional space according to the final ranging result.

[0063] The purpose of this step is to further extract the real-time position and dynamic movement state information of the target in three-dimensional space after completing the unified spatial alignment and Doppler motion compensation of the lidar ranging data, providing basic data support for subsequent target behavior analysis, navigation decision-making, or early warning control.

[0064] Specifically, outputting the accurate position and movement state information of the target in three-dimensional space according to the final ranging result includes the following steps: S61: Obtain the three-dimensional coordinate values of the target mapped to the end of the dynamic ranging reference coordinate system ; It should be noted that the target three-dimensional coordinates refer to the spatial position coordinates corresponding to the target after lidar ranging, dynamic compensation, and coordinate transformation processing at the current moment, in the form of a three-dimensional vector (X, Y, Z). By invoking the dynamic ranging reference coordinate system constructed in step S3, the Doppler-corrected distance result at the current moment can be coordinate-transformed with the platform attitude position information, so as to map the target position to a unified three-dimensional coordinate system, ensure that all target data is under the same spatial reference, and have good spatial consistency and comparability.

[0065] S62: Combine the ranging results of multiple consecutive moments to construct a target space trajectory sequence.

[0066] This step aims to mine the motion trajectory characteristics of the target through time series information. Specifically, the system caches the ranging data of several consecutive time frames (for example, within the past 1-2 seconds), extracts and records the target three-dimensional position coordinate values at the corresponding moments, and sequentially forms the position trajectory sequence of the target under continuous time, expressed as: , ,..., . Among them, is the target spatial position at the i-th moment, and this trajectory sequence provides an input basis for the extraction of subsequent dynamic state quantities such as velocity and acceleration.

[0067] It should be noted that this trajectory sequence has a timestamp identifier during the acquisition process, so as to support differential calculation based on the time dimension, and optionally combine mechanisms such as sliding window and recursive cache to optimize the data processing efficiency.

[0068] S63: Based on the target space trajectory sequence, calculate the dynamic motion state information of the target and output the real-time position and motion state of the target in three-dimensional space, so as to accurately depict the motion behavior of the target and its dynamic distribution characteristics in three-dimensional space.

[0069] Among them, the dynamic motion state information includes velocity vector, acceleration vector, and motion direction.

[0070] Specifically, the calculation of the motion state information is based on the target ranging results of multiple consecutive moments. Through the position difference method, Kalman filter or other state estimation algorithms, the velocity change trend and acceleration characteristics of the target are extracted to reflect the motion behavior of the target in real time. The calculation method is as follows: Velocity , where: and are the spatial position coordinates of the target in the current frame and the previous frame, and is the corresponding timestamp, and the obtained result is the three-dimensional velocity vector of the target at the current moment, with the unit of m / s.

[0071] Acceleration , where: and are the velocity vectors of the target in the current frame and the previous frame, with the unit of .

[0072] This acceleration vector reflects the change trend of the target's velocity and is used to determine whether the target is in an accelerating, decelerating, or uniform motion state.

[0073] In summary, by introducing the Doppler frequency shift feature extraction and frequency shift resolution model, the present invention calculates the relative velocity of the target and performs dynamic motion compensation on the lidar ranging result, enabling the ranging process to adapt to the ranging error caused by the high-speed movement of the target. Compared with the traditional ranging methods based on static or uniform models, the present invention can effectively solve the ranging deviation problem caused by high-speed targets and significantly improve the ranging accuracy and stability in long-distance dynamic scenarios. The present invention fuses the data of IMU, GPS, and visual SLAM image coordinate systems to construct a dynamic ranging reference coordinate system, enabling the lidar ranging result to be aligned with the spatial state change of the test platform in real time. Compared with the traditional methods relying on a fixed reference system, the present invention has higher environmental adaptability and robustness, can effectively cope with the coordinate system perturbation problems during complex dynamic processes such as the movement, turning, and vibration of the test platform, and ensure the spatial consistency and time synchronization of the ranging data. The present invention not only outputs the position coordinates of a long-distance target in three-dimensional space but also further calculates the dynamic information such as the velocity and acceleration of the target based on the ranging sequences at multiple moments, enabling the functions of target behavior analysis and trajectory tracking. This extended ability endows the present invention with the potential to expand to higher-order application scenarios such as target recognition, motion prediction, and dynamic obstacle detection, significantly improving the practical value of the lidar system in intelligent transportation, dynamic monitoring, and environmental perception.

[0074] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered within the scope of the claims of the present invention.

Claims

1. A laser radar test method based on long-distance target ranging results, characterized in that: The following steps are involved: Acquire a laser radar echo signal and extract spectrum information of the echo signal; The Doppler frequency shift characteristic parameters are extracted according to the spectrum information, and the relative speed of the target is calculated based on a preset frequency shift solution model. ; Obtain multi-source sensor data associated with the LiDAR test platform and fuse them to construct a dynamic ranging reference coordinate system; Performing dynamic motion compensation on the laser ranging result according to the relative speed of the target to obtain the initial measured distance of the target after Doppler correction; Mapping the Doppler-corrected target initial distance measurement to the dynamic distance measurement reference coordinate system to obtain a long-distance target final distance measurement result aligned with the test platform coordinate system; The accurate position and movement status information of the target in three-dimensional space are output according to the final ranging result.

2. The laser radar testing method based on the long-distance target ranging result as claimed in claim 1, characterized in that: The frequency shift solution model is based on the laser wavelength and frequency shift value , the relative speed is calculated using the following relationship: ， in, is the relative speed of the target, and the obtained value is used for dynamic ranging compensation. is the Doppler frequency shift, c is the speed of light, is the emission frequency of the laser.

3. The laser radar testing method based on the long-distance target ranging result as claimed in claim 1, characterized in that: The multi-source sensor data includes but is not limited to IMU data, GPS data, and visual sensor image data.

4. The laser radar testing method based on the long-distance target ranging result as claimed in claim 3, characterized in that: Acquiring the multi-source sensor data associated with the laser radar test platform and fusing them to construct a dynamic ranging reference coordinate system includes the following steps: Calculate the attitude change information of the test platform according to the IMU data; Acquire the location information of the test platform according to the GPS data; Constructing a visual SLAM image coordinate system of the environment in which the platform is located according to the visual sensor image data; Based on the fusion of the IMU data, the GPS data and the visual SLAM image coordinate system data, a dynamic ranging reference coordinate system is constructed.

5. The laser radar testing method based on the long-distance target ranging result as claimed in claim 4, characterized in that: Based on the fusion of the IMU data, the GPS data and the visual SLAM image coordinate system data, a dynamic ranging reference coordinate system is constructed, including the following steps: The visual SLAM image coordinate system is used as the initial environment reference system, the posture change information derived by the IMU is used as the intermediate state constraint, and the position information of the test platform provided by the GPS is used as the global position factor to construct a fusion optimization model, wherein the fusion optimization model is a factor graph model; A nonlinear optimization algorithm is used to globally optimize the above factor graph to obtain the optimal state estimation results of the platform at each moment, including the three-dimensional position coordinates (X, Y, Z) and the three-axis attitude angle; The current platform state output by factor graph optimization is used as the coordinate system reference to obtain the three-dimensional dynamic ranging reference coordinate system.

6. The laser radar testing method based on long-distance target ranging results as claimed in claim 1, characterized in that: Dynamic motion compensation is performed on the laser ranging result according to the relative speed of the target to obtain the initial measured distance of the target after Doppler correction, including the following steps: Get the original laser radar ranging results , and determine the data format and unit of the original ranging result; The relative speed of the target is calculated , determine the time correction required to compensate ; According to the compensation time correction amount , combined with the relative speed of the target , calculate the distance of the ranging offset caused by the target motion ; Offset the distance Compared with the original distance measurement result Perform difference calculation to obtain the corrected distance after Doppler motion compensation , the specific calculation formula is: ; Corrected distance after Doppler motion compensation Perform filtering and smoothing processing to output the initial target distance after Doppler correction.

7. The laser radar testing method based on the long-distance target ranging result as claimed in claim 6, characterized in that: According to the calculated relative speed of the target , determine the time correction required to compensate , including the following steps: According to the speed of light constant c, calculate the one-way propagation time of the laser signal , the propagation time Used to estimate the actual displacement of the target during the laser propagation period; The relative speed of the target One-way propagation time Correlation, determining the displacement of the target during the signal propagation , the formula is: , the displacement Reflects the spatial displacement of the target caused by its own motion during the round trip of the laser; The displacement Convert to the corresponding time correction .

8. The laser radar testing method based on long-distance target ranging results as claimed in claim 1, characterized in that: Mapping the Doppler-corrected target initial distance to the dynamic distance measurement reference coordinate system to obtain a final distance measurement result of the long-distance target aligned with the test platform coordinate system includes the following steps: Get the initial distance of the target after Doppler correction Data, and determine the data format and coordinate reference system attributes of the target's initial distance measurement; Obtain the dynamic ranging reference coordinate system information output by factor graph optimization at the current moment, including the platform's three-dimensional position and attitude angle; According to the installation posture parameters of the laser radar on the test platform, the posture transformation matrix between the local coordinate system to which the original ranging data belongs and the platform coordinate system is constructed; The initial measured distance data of the target is converted into the three-dimensional position coordinates of the target in the local coordinate system of the test platform through the coordinate transformation matrix; According to the current global state information of the test platform, the three-dimensional position coordinates of the target in the local coordinate system of the platform are mapped to the dynamic ranging reference coordinate system; The three-dimensional coordinate values ​​of the target that has completed spatial alignment are output as the final ranging result of the long-distance target aligned with the test platform coordinate system.

9. The laser radar testing method based on the long-distance target ranging result as claimed in claim 8, characterized in that: Outputting the accurate position and movement status information of the target in three-dimensional space according to the final ranging result includes the following steps: Get the target 3D coordinates mapped to the dynamic ranging reference coordinate system ; Combining the ranging results of multiple consecutive moments, constructing the target space trajectory sequence; Based on the target space trajectory sequence, the dynamic motion state information of the target is calculated and the real-time position and motion state of the target in three-dimensional space are output; The dynamic motion state information includes a velocity vector, an acceleration vector and a motion direction.

10. The laser radar testing method based on long-distance target ranging results according to claim 9, characterized in that: The calculation of the dynamic motion state information is based on the target ranging results at multiple consecutive moments. The speed change trend and acceleration characteristics of the target are extracted through position difference method, Kalman filtering or other state estimation algorithms to reflect the target's motion behavior in real time.

Images