IMU (Inertial Measurement Unit) and laser radar fused three-dimensional point cloud motion distortion removal method

Through the fusion of IMU and lidar, the real-time correction of LIDAR point clouds is used to use IMU data to solve the distortion problem caused by sensor movement, improve the accuracy and consistency of point cloud data, and is suitable for autonomous driving applications in complex dynamic environments.

CN119048404APending Publication Date: 2024-11-29TIANJIN UNIV
View PDF 7 Cites 0 Cited by

Patent Information

Application Number
CN202410984572.1
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-22
Publication Date
2024-11-29

AI Technical Summary

Technical Problem

In autonomous driving and robot navigation, the three-dimensional point cloud data collected by lidar is susceptible to distortion caused by sensor motion, resulting in reduced data accuracy and consistency. The existing technology has failed to fully utilize the dynamic motion information provided by IMU for real-time correction. Especially in complex dynamic environments, there are significant limitations.

Method used

Through the fusion of IMU and lidar, the three-axis acceleration and angular velocity data are obtained, queue synchronization processing is performed, the data is aligned and the LIDAR point cloud is corrected in real time using IMU data. The external parameters are optimized by the least squares method to realize the three-dimensional point cloud data. Motion distortion compensation.

Benefits of technology

It significantly improves the accuracy and consistency of point cloud data, is suitable for complex dynamic environments such as mining areas and factory operation scenarios, and improves the data processing reliability of autonomous driving vehicles.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119048404A_ABST
    Figure CN119048404A_ABST
Patent Text Reader

Abstract

The invention discloses an IMU (Inertial Measurement Unit) and laser radar fused three-dimensional point cloud motion distortion removing method, which comprises the following steps of: 1, acquiring three-axis acceleration data (accx, accy and accz) and three-axis angular rate data (Gyrox, Gyroy and Gyroz) from an IMU sensor, and acquiring three-dimensional point cloud data (x, y, z, institution, ring and time) from a laser radar sensor; step 2, calibrating a laser radar sensor and an IMU sensor to the same coordinate system; 3, aligning the IMU data and the LIDAR data after coordinate system conversion by adopting a queue synchronization method, so that the IMU data and the LIDAR data are processed under the same timestamp; and 4, after the IMU data and the LIDAR data are aligned, performing motion distortion compensation on the LIDAR original point cloud by using the three-axis acceleration data and the three-axis angular rate data provided by the IMU, and performing real-time correction on the LIDAR original point cloud data according to the motion state of the vehicle. According to the method, the precision and consistency of the point cloud data are remarkably improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of autonomous driving technology, and in particular to a three-dimensional point cloud motion dedistortion method that fuses an IMU (inertial measurement unit) and a laser radar (LIDAR). The method combines the three-axis acceleration and three-axis angle data provided by the IMU with the three-dimensional point cloud data collected by the laser radar, thereby effectively correcting the point cloud distortion caused by sensor movement. Background Art

[0002] In applications such as autonomous driving and robot navigation, laser radar (LIDAR) and inertial measurement unit (IMU) sensors are key perception devices. LIDAR sensors can capture three-dimensional point cloud data of the environment with high accuracy, while IMU sensors provide three-axis acceleration and angular velocity information about the motion state of the device. However, since the vehicle or device is in motion, the point cloud data collected by LIDAR may be affected by the distortion caused by the movement of the sensor, resulting in reduced accuracy and consistency of the point cloud data.

[0003] In traditional methods, point cloud data processing usually relies on data from a single sensor, which makes it difficult to cope with complex dynamic environmental changes. Existing point cloud dedistortion methods often ignore the advantages of sensor fusion and do not fully utilize the dynamic motion information provided by the IMU to perform real-time correction of the LIDAR point cloud. This processing method has obvious limitations in practical applications, especially for autonomous driving projects in mining areas and engineering autonomous driving vehicles such as forklifts operating in factory areas.

[0004] In order to improve the accuracy and consistency of point cloud data, this patent proposes a variety of sensor fusion methods. By fusing the three-axis acceleration and three-axis angular velocity data provided by the IMU with the three-dimensional point cloud data collected by the LIDAR, the point cloud distortion caused by sensor movement can be effectively corrected. However, in current research and applications, there are still many technical difficulties in how to efficiently perform this data fusion and how to achieve real-time distortion compensation in a dynamic environment.

[0005] Existing studies have shown that the motion state of a vehicle has a significant impact on the accuracy of point cloud data. For example, in mining operations with many potholes or in forklift operations in factories, the dynamic motion of the vehicle, such as acceleration, deceleration, and turning, can cause significant distortion of point cloud data. Traditional point cloud processing methods based on static models cannot adapt to such dynamic changes, resulting in reduced reliability of data processing results.

[0006] Therefore, it is particularly important that this patent proposes a 3D point cloud motion dedistortion algorithm based on the fusion of IMU and LiDAR. By integrating the dynamic motion information provided by IMU, real-time correction of LIDAR point cloud data can be achieved, significantly improving the accuracy and consistency of point cloud data. This is crucial for the application of autonomous driving in engineering projects such as mining trucks or unmanned forklifts. Summary of the invention

[0007] The purpose of the present invention is to provide a three-dimensional point cloud motion dedistortion method by fusing IMU and LiDAR to address the problem that point cloud data in the prior art is prone to significant distortion.

[0008] The technical solution adopted to achieve the purpose of the present invention is:

[0009] A three-dimensional point cloud motion dedistortion method based on IMU and laser radar fusion includes the following steps:

[0010] Step 1: Get the three-axis acceleration data from the IMU sensor (acc x ,acc y ,acc z ) and three-axis angular rate data (Gyro x , Gyro y , Gyro z ), obtain 3D point cloud data (x, y, z, intensity, ring, time) from the lidar sensor;

[0011] Step 2: calibrate the lidar sensor and IMU sensor to the same coordinate system;

[0012] Step 3: Use the queue synchronization method to align the IMU data and LIDAR data after coordinate system conversion so that they can be processed at the same timestamp;

[0013] Step 4: After the IMU data and LIDAR data are aligned, the three-axis acceleration data and three-axis angular rate data provided by the IMU are used to compensate for the motion distortion of the LIDAR original point cloud, and the LIDAR original point cloud data is corrected in real time according to the vehicle's motion state.

[0014] In the above technical solution, the three-dimensional point cloud motion dedistortion method also includes step 5, publishing the corrected and registered point cloud data and displaying it on the vehicle-machine interaction platform.

[0015] In the above technical solution, in step 1, the ROS communication mechanism under the UBUNTU system is used to realize the transmission, reception and fusion of three-axis acceleration data, three-axis angular rate data and three-dimensional point cloud data.

[0016] In the above technical solution, the calibration method in step 2 comprises the following steps:

[0017] Let the point in the IMU coordinate system be P IMU , the point in the LiDAR coordinate system is P LiDAR , the relationship between the two is: P LiDAR =RP IMU +t, where: R is a 3×3 rotation matrix used to describe the rotation relationship from the IMU to the LIDAR coordinate system, t is a 3×1 translation vector used to describe the translation relationship from the IMU coordinate system to the LIDAR coordinate system, and the external parameters include R and t.

[0018] In the above technical solution, the least square method is used to further optimize the external parameters R and t:

[0019]

[0020] P IMU,i is the point in the IMU coordinate system at time i, P LiDAR,i is the point in the LIDAR coordinate system at time i.

[0021] In the above technical solution, in step 3, it is assumed that the IMU data queue is Q IMU , the LiDAR data queue is Q LiDAR , for each frame of LiDAR data P LiDAR,i , whose timestamp is t LiDAR,i , from the IMU queue Q IMU Find the LiDAR,i Recent IMU data t IMU,j , and then determine the timestamp t of the IMU data IMU,j Is it within the allowed time range:

[0022] |t LiDAR,i -t IMU,j |≤Δt max

[0023] Among them, Δt max is the preset maximum tolerable time difference;

[0024] If the timestamp of the IMU data is t IMU,j The timestamp t of the LiDAR data LiDAR,i Not completely aligned, use linear interpolation to calculate IMU data at t LiDAR,i Estimated value of the moment.

[0025] In the above technical solution, the linear interpolation formula is as follows:

[0026]

[0027] Among them, P IMU,interp is the IMU data at t LiDAR,i The estimated value of the moment, P IMU,j The timestamp is t IMU,j IMU data, t LiDAR,i is the timestamp of the LiDAR data at time i, t IMU,j is the timestamp of the IMU data at time j, t IMU,j+1 To follow t IMU,j The timestamp after that, P IMU,j+1 is the timestamp immediately following t IMU,j IMU data afterwards.

[0028] In the above technical solution, in step 4, assuming that the linear velocity and angular velocity of the vehicle remain unchanged within the time interval Δt, the movement of the point cloud data point is expressed as:

[0029] Δx=v·Δt·cos(θ)

[0030] Δy=v·Δt·sin(θ)

[0031]

[0032] Among them, Δx, Δy, Δz are the displacement compensation of the point cloud data point in the x, y, and z axis directions respectively, Δt is the time interval, and θ is the initial angle of the point cloud data point, which is calculated by the initial position of the point cloud data point. is the pitch angle of the vehicle during operation, x, y, z are the initial position coordinates of the point cloud data points, v is the linear speed of the vehicle,

[0033] In the above technical solution, the initial position of the point cloud data point and the motion compensation are combined to obtain the compensated new position: the position compensation of the point cloud data point is expressed as:

[0034]

[0035] Among them, x, y, z are the initial coordinates of the point cloud data point, x′, y′, z′ are the new coordinates of the point cloud data point after compensation, Δx, Δy, Δz are the displacement compensation amounts of the point cloud data point in the x, y, z axis directions respectively;

[0036] In the above technical solution, the rotation matrix R′ is calculated according to the angular velocity data provided by the IMU:

[0037]

[0038] In the above technical solution, the angle change of the point cloud data point is as follows:

[0039] Δθ=ω·Δt

[0040] Where ω is the angular velocity of the vehicle,

[0041] Compared with the prior art, the present invention has the following beneficial effects:

[0042] 1. Compared with the traditional motion distortion compensation algorithm, the present invention eliminates point cloud motion distortion in three dimensions for working scenes of engineering vehicles in harsh working environments such as mining areas or factories, and performs real-time correction of LIDAR point cloud data, significantly improving the accuracy and consistency of point cloud data.

[0043] 2. The algorithm of the present invention has a clear flow, a simple structure, does not require a large amount of computing power, and is easy to implement. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 It is the overall framework diagram of the algorithm of the present invention. DETAILED DESCRIPTION

[0045] The present invention is further described in detail below in conjunction with specific embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0046] Example 1

[0047] A three-dimensional point cloud motion dedistortion method based on IMU and laser radar fusion includes the following steps:

[0048] Step 1: Get the three-axis acceleration (acc) from the IMU sensor x ,acc y ,acc z ) and three-axis angular rate (Gyro x , Gyro y , Gyro z ) data. acc x ,acc y ,acc z Gyro is the acceleration of the vehicle in the x, y, and z directions respectively. x , Gyro y , Gyro z The angular velocity of the vehicle in the x, y, and z directions is obtained from the lidar sensor. The data is parsed and stored through the message subscription mechanism to prepare for subsequent processing.

[0049] More specifically, the ROS communication mechanism under the UBUNTU system can realize the reception and fusion of three-axis acceleration, three-axis angular rate data and three-dimensional point cloud data. This communication mechanism is mostly used in unmanned driving systems and robot operating systems.

[0050] Step 2: Convert the coordinate systems of the LiDAR sensor and the IMU sensor to the same coordinate system through calibration or sensor translation and rotation.

[0051] More specifically, the two sensors are calibrated to the same coordinate system through calibration software or by modifying the parameters of the LiDAR driver and IMU external calibration software. The principle of calibrating the two sensors to the same coordinate system through calibration software is as follows:

[0052] Let the point in the IMU coordinate system be P IMU , the point in the LiDAR coordinate system is P LiDAR The relationship between the two can be expressed as: LiDAR =RP IMU +t, where: R is a 3×3 rotation matrix used to describe the rotation relationship from the IMU to the LIDAR coordinate system. t is a 3×1 translation vector used to describe the translation relationship from the IMU coordinate system to the LIDAR coordinate system.

[0053] The rotation matrix R can be obtained by quaternion (q w ,q x ,q y ,q z ) indicates that the conversion relationship between quaternion and rotation matrix is:

[0054]

[0055] The ICP algorithm is used to accurately align the point cloud data of IMU and LIDAR. The formula is:

[0056]

[0057] Where: P IMU,i is a point in the IMU coordinate system. LiDAR,i is a point in the LiDAR coordinate system. The least squares method is used to further optimize the external parameters, and the formula is:

[0058]

[0059] Through the above process and formula, the external parameter calibration of IMU and LiDAR can be realized to ensure that the data of the two sensors are fused and processed in the same coordinate system, thereby improving the accuracy and consistency of the point cloud data.

[0060] Step 3: When fusing IMU and LIDAR data, it is crucial to ensure that the two sensor data are aligned at the same timestamp. In order to ensure the synchronization of the timestamps of the LiDAR sensor and the IMU sensor, the queue synchronization method is used to align the IMU data and the LIDAR data so that they can be processed at the same timestamp. The specific steps are as follows:

[0061] Store the IMU data and LiDAR data in their respective queues in the order of timestamps. Assume that the IMU data queue is Q IMU , the LiDAR data queue is Q LiDAR For each frame of LiDAR data P LiDAR,i (Time stamp is t LiDAR,i ), from the IMU queue Q IMU Find the LiDAR,i Recent IMU data t IMU,j , and then determine the timestamp t of the IMU data IMU,j Is it within the allowed time range:

[0062] |t LiDAR,i -t IMU,j |≤Δt max

[0063] Among them, Δt max It is the preset maximum tolerable time difference.

[0064] If the timestamp of the IMU data is t IMU,j The timestamp t of the LiDAR data LiDAR,i Not completely aligned, use linear interpolation to calculate IMU data at t LiDAR,i The estimated value at the moment. The linear interpolation formula is as follows:

[0065]

[0066] Among them, P IMU,interp is the IMU data at t LiDAR,i The estimated value of the moment, P IMU,j The timestamp is t IMU,j IMU data, t LIDAR,i is the timestamp of the LiDAR data at time i, t IMU,j is the timestamp of the IMU data at time j, t IMU,j+1 To follow t IMU,j The timestamp after that, P IMU,j+1 is the timestamp immediately following t IMU,j IMU data afterwards.

[0067] Step 4: Use the three-axis acceleration and angular velocity change rate data provided by the IMU to compensate for motion distortion of the LIDAR original point cloud. According to the vehicle's motion state, the point cloud data is corrected in real time.

[0068] Based on the three-axis acceleration data provided by the IMU, the linear velocity of the vehicle can be calculated:

[0069]

[0070] v is the linear velocity of the vehicle.

[0071] According to the three-axis angular velocity data provided by the IMU, the angular velocity of the vehicle can be calculated:

[0072]

[0073] ω is the angular velocity of the vehicle.

[0074] Assuming that the linear velocity and angular velocity of the vehicle remain constant within the time interval Δt, the motion of the point cloud data points can be expressed as:

[0075] Δx=v·Δt·cos(θ)

[0076] Δy=v·Δt·sin(θ)

[0077]

[0078] Among them, Δx, Δy, and Δz are the displacement compensation amounts of the point cloud data points in the x, y, and z axis directions, respectively. and x, y, z are the initial position coordinates of the point cloud data points;

[0079] The angle change of the point cloud data point is shown in the following formula:

[0080] Δθ=ω·Δt

[0081] Combine the initial position of the point cloud data point with the motion compensation to get the compensated new position:

[0082]

[0083] Among them, x, y, z are the initial coordinates of the point cloud data point, x′, y′, z′ are the compensated new coordinates of the point cloud data point, Δx, Δy, Δz are the displacement compensation amounts of the point cloud data point in the x, y, z axis directions, respectively.

[0084] Considering the point cloud motion distortion compensation in three-dimensional space, the rotation matrix is ​​calculated based on the angular velocity data provided by the IMU:

[0085]

[0086] Step 5: Publish the corrected and registered point cloud data and display it on the vehicle-machine interaction platform for subsequent processing by algorithm development and debugging personnel.

[0087] Example 2

[0088] A three-dimensional point cloud motion dedistortion method based on IMU and laser radar fusion includes the following steps:

[0089] Step 1: Install the driver for the laser radar point cloud and obtain the raw data of the laser radar point cloud, whose data format is (x, y, z, intensity, ring, timestamp); the laser radar used in this embodiment is the Suteng 128-line mechanical laser radar; the IMU used is the POMS-G6615 model combined inertial navigation; in the UBUNTU system, through the serial port connection, parse the x, y, z axis three-axis acceleration information and three-axis angular velocity information we need, and send them in the form of ROS communication;

[0090] Step 2, according to the installation position of the IMU and the installation position of the laser radar; perform coordinate system conversion between the two sensors; in this embodiment, through the coordinate system conversion work, the coordinate system of the combined inertial navigation is calibrated to the position of the laser radar coordinate system through translation and rotation transformation; its translation vector is:

[0091] [-0.897, +1.812, -0.562]

[0092] When sending the original point cloud of the LiDAR, the coordinate system and direction of the LiDAR are unified with the direction of the IMU coordinate system, both of which are right-handed systems; therefore, the quaternion and Euler angle are as follows:

[0093] [1, 0, 0, 0]

[0094] [0, 0, 0]

[0095] Step 3, synchronize the timestamps of IMU messages and lidar point cloud messages. First, I defined two queues in the code, one for storing IMU data and the other for storing lidar data. Then, in the IMU callback and the lidar point cloud callback, the parsed IMU data and its timestamp are stored in the IMU queue, and the point cloud data and its timestamp are stored in the lidar queue. Finally, the queue messages are processed in an independent function to synchronize the timestamps.

[0096] Step 4 is the core step of laser radar point cloud motion dedistortion: First, receive the laser radar point cloud data and store it in the member variables of the class. Next, calculate the time difference (Δt) between the current point cloud data and the previous point cloud data. If the rotation frequency is set, use its reciprocal as the time difference; otherwise, calculate the time difference based on the timestamp. Initialize the static variables that store the linear velocity and angular velocity to calculate the average velocity. Convert the point cloud data from the ROS message format to the PCL point cloud format for subsequent processing. Read the width and height of the point cloud for traversing the point cloud data.

[0097] Calculate the linear velocity and angular velocity of the vehicle based on the IMU data, and retain the value of the previous moment to calculate the average velocity. Calculate the average linear velocity and angular velocity of the current moment and the previous moment. Process each point in the point cloud: read the coordinates of the point, calculate the angle θ, calculate the angle change Δθ based on the angular velocity, calculate the position correction x′, y′, z′ based on the linear velocity, perform rotation correction on the point, and overwrite the original value with the corrected value.

[0098] Step 5, convert the corrected point cloud data from PCL format back to ROS message format and publish it to the specified topic. For debugging and verification, print the coordinates of the first corrected point. Through these steps, the lidar point cloud data can be effectively processed to eliminate the distortion caused by vehicle motion, thereby improving the accuracy and consistency of the point cloud data. These steps provide a reliable foundation for data processing in multi-sensor fusion and autonomous driving systems.

[0099] The above is only a preferred embodiment of the present invention. It should be pointed out that, for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A three-dimensional point cloud motion dedistortion method based on IMU and laser radar fusion, characterized in that: The following steps are involved: Step 1: Get the three-axis acceleration data (acc x ,acc y ,acc z ) and three-axis angular rate data (Gyro x , Gyro y , Gyro z ), obtain 3D point cloud data (x, y, z, intensity, ring, time) from the lidar sensor; Step 2: calibrate the lidar sensor and IMU sensor to the same coordinate system; Step 3: Use the queue synchronization method to align the IMU data and LIDAR data after coordinate system conversion so that they can be processed at the same timestamp; Step 4: After the IMU data and LIDAR data are aligned, the three-axis acceleration data and three-axis angular rate data provided by the IMU are used to compensate for the motion distortion of the LIDAR original point cloud, and the LIDAR original point cloud data is corrected in real time according to the vehicle's motion state.

2. The method for 3D point cloud motion dedistortion by integrating IMU and laser radar as claimed in claim 1, characterized in that: The three-dimensional point cloud motion dedistortion method also includes step 5, publishing the corrected and registered point cloud data and displaying it on the vehicle-machine interaction platform.

3. The method for 3D point cloud motion dedistortion by integrating IMU and laser radar as claimed in claim 1, characterized in that: In step 1, the ROS communication mechanism under the UBUNTU system is used to realize the reception, transmission and fusion of three-axis acceleration data, three-axis angular rate data and three-dimensional point cloud data.

4. The method for 3D point cloud motion dedistortion by integrating IMU and laser radar as claimed in claim 1, characterized in that: The calibration method in step 2 comprises the following steps: Let the point in the IMU coordinate system be P IMU , the point in the LiDAR coordinate system is PLi DAR , the relationship between the two is: P LiDAR =RP1 MU +t, where: R is a 3×3 rotation matrix used to describe the rotation relationship from the IMU to the LIDAR coordinate system, t is a 3×1 translation vector used to describe the translation relationship from the IMU coordinate system to the LIDAR coordinate system, and the external parameters include R and t.

5. The method for 3D point cloud motion dedistortion by integrating IMU and laser radar as claimed in claim 4, characterized in that: The least squares method is used to further optimize the external parameters R and t: P IMU,i is the point in the IMU coordinate system at time i, P LiDAR,i is the point in the LIDAR coordinate system at time i.

6. The method for 3D point cloud motion dedistortion by integrating IMU and laser radar as claimed in claim 1, characterized in that: In step 3, it is assumed that the IMU data queue is Q IMU , the LiDAR data queue is Q LiDAR , for each frame of LiDAR data P LiDAR,i , whose timestamp is t LiDAR,i , from the IMU queue Q IMU Find the value corresponding to t LiDAR,i Recent IMU data t IMU,j , and then determine the timestamp t of the IMU data IMU,j Is it within the allowed time range: |t LiDAR,i -t IMU,j |≤Δt max Among them, Δt max is the preset maximum tolerable time difference; If the timestamp of the IMU data is t IMU,j The timestamp t of the LiDAR data LiDAR,i Not completely aligned, use linear interpolation to calculate IMU data at t LiDAR,i Estimated value of the moment.

7. The method for 3D point cloud motion dedistortion by integrating IMU and laser radar as claimed in claim 6, characterized in that: The linear interpolation formula is as follows: Among them, P IMU,interp is the IMU data at t LiDAR,i The estimated value of the moment, P IMU,j The timestamp is t IMU,j IMU data, t LiDAR,i is the timestamp of the LiDAR data at time i, t IMU,j is the timestamp of the IMU data at time j, t IMU,j+1 To follow t IMU,j The timestamp after that, P IMU,j+1 is the timestamp immediately following t IMU,j IMU data afterwards.

8. The method for 3D point cloud motion dedistortion by integrating IMU and laser radar as claimed in claim 1, characterized in that: In step 4, assuming that the linear velocity and angular velocity of the vehicle remain unchanged within the time interval Δt, the movement of the point cloud data points is expressed as: Δx=v·Δt·cos(θ) Δy=v·Δt·sin(θ) Among them, Δx, Δy, and Δz are the displacement compensation amounts of the point cloud data points in the x, y, and z axis directions, respectively. θ is the initial angle of the point cloud data point, which is calculated based on the initial position of the point cloud data point. is the pitch angle of the vehicle during operation, x, y, z are the initial position coordinates of the point cloud data points, v is the linear speed of the vehicle, 9. The method for 3D point cloud motion dedistortion by integrating IMU and laser radar as claimed in claim 8, characterized in that: The initial position of the point cloud data point and the motion compensation are combined to obtain the compensated new position: The position compensation of the point cloud data point is expressed as: Among them, x, y, z are the initial coordinates of the point cloud data point, x′, y′, z′ are the compensated new coordinates of the point cloud data point, Δx, Δy, Δz are the displacement compensation amounts of the point cloud data point in the x, y, z axis directions, respectively.

10. The method for 3D point cloud motion dedistortion by integrating IMU and laser radar as claimed in claim 9, characterized in that: According to the angular velocity data provided by the IMU, calculate the rotation matrix R′: In the above technical solution, the angle change of the point cloud data point is as follows: Δθ=ω·Δt Where ω is the angular velocity of the vehicle,

Citation Information

Patent Citations

  • Method for correcting multi-line laser radar point cloud movement distortion on the basis of multi-sensor fusion

    CN109975792A

  • Laser point cloud data distortion removing method based on kernel ridge regression

    CN116167919A

  • External parameter calibration method, device and equipment for laser radar and inertial navigation and medium

    CN117388830A

  • IMU-fused three-dimensional laser radar positioning and mapping method

    CN117419719A

  • Laser point cloud distortion removing method and device under fast motion and storage medium

    CN118226419A