An online high-precision map building method and system based on positioning fusion

Abstract

This invention discloses an online high-precision map building method and system based on positioning fusion, including: acquiring three-axis acceleration data and three-axis gyroscope data through an IMU; obtaining predicted continuous attitude change data based on the changes in velocity and angle; matching and building a local map through LiDAR point cloud data, and acquiring the driving trajectory; matching the driving trajectory with the local map to obtain LiDAR observation pose data; obtaining satellite positioning trajectory data fed back by RTK; obtaining the overlap degree in a unified coordinate system by fitting the predicted continuous attitude change data, LiDAR observation pose data, and satellite positioning trajectory data at the same time, and obtaining the variance value of the relative deviation; judging the positioning accuracy of the predicted continuous attitude change data, LiDAR observation pose data, and satellite positioning trajectory data at the same time based on the obtained variance value of the relative deviation, and obtaining the pose of the sub-map at the same moment.

Description

Technical Field

[0001] This invention relates to the field of autonomous driving, specifically to an online high-precision map building method and system based on positioning fusion. Background Technology

[0002] Traditional RTK positioning for autonomous driving suffers from decreased positioning accuracy in areas with poor RTK signal, failing to meet the requirements of autonomous driving. Furthermore, existing high-precision mapping technologies have drawbacks, including the lack of real-time mapping capabilities, the need for offline mapping via a backend server, the requirement for manual post-processing and correction of map accuracy, and the inability of high-precision point cloud maps to correspond to the actual latitude and longitude coordinate system.

[0003] Therefore, how to automatically generate high-precision maps that meet latitude and longitude coordinates is a topic that industry researchers need to study. Summary of the Invention

[0004] The purpose of this invention is to overcome the shortcomings of the prior art and provide an online high-precision map building method based on location fusion, comprising the following steps:

[0005] The three-axis acceleration data and three-axis gyroscope data are acquired through the IMU. The changes in velocity and angle are obtained from the three-axis acceleration data and three-axis gyroscope data, respectively. Based on the changes in velocity and angle, the predicted continuous attitude change data is obtained.

[0006] The local map is built by matching the lidar point cloud and the driving trajectory is obtained. The driving trajectory is matched with the local map to obtain lidar observation pose data.

[0007] Obtain satellite positioning trajectory data fed back by RTK;

[0008] By fitting the predicted continuous attitude change data, lidar observation pose data, and satellite positioning trajectory data at the same moment, the degree of overlap in a unified coordinate system is obtained, and the variance of the relative deviation is obtained.

[0009] The positioning accuracy of the predicted continuous attitude change data, lidar observed pose data, and satellite positioning trajectory data at the same moment is judged by the variance value of the obtained relative deviation, so as to obtain the pose of the sub-map at the same moment.

[0010] Furthermore, the method of obtaining the changes in velocity and angle using triaxial acceleration data and triaxial gyroscope data respectively includes:

[0011] The change in velocity is obtained by integrating the triaxial acceleration data, and the change in angle is obtained by integrating the angular velocity obtained from the triaxial gyroscope data. The change in displacement of the vehicle in space is obtained by combining the change in angle and the change in velocity.

[0012] Furthermore, the lidar observation pose data includes three-dimensional spatial coordinates and three-dimensional rotation angle pose data.

[0013] Furthermore, it also includes converting the predicted continuous attitude change data and lidar observed pose data at the same time moment into latitude and longitude coordinates respectively.

[0014] Furthermore, the method of obtaining the degree of overlap in a unified coordinate system through fitting includes:

[0015] By locally optimizing the origin coordinates of the three space curves, the spatial distance error at the same coordinate at each moment is minimized.

[0016] Furthermore, obtaining the variance value of the relative deviation includes:

[0017] By optimizing the fit of three curves, the expected value of each point on the three curves is obtained. Then, the expected value of the square of the deviation between all points on each curve and the expected value of the corresponding point is calculated.

[0018] Furthermore, the method of judging the positioning accuracy of the predicted continuous attitude change data, lidar observed pose data, and satellite positioning trajectory data at the same moment by using the variance value of the obtained relative deviation to obtain the pose of the sub-map at the same moment includes:

[0019] When the RTK and SLAM localizations are consistent, the pose of the submap is determined by finding the minimum deviation between the two. When the RTK and SLAM localizations are inconsistent, the pose of the submap is determined by the constraint relationships between the point cloud and the submap, and between submaps, and the corresponding calculated observation variance matrix.

[0020] An online high-precision map building system based on positioning fusion, which applies the aforementioned method, includes a data acquisition module, a satellite positioning module, a data processing module, a map building module, and a communication device; the data acquisition module, satellite positioning module, map building module, and communication device are respectively connected to the data processing module.

[0021] Preferably, the data acquisition module includes a three-axis acceleration data acquisition device, a three-axis gyroscope module, and a lidar point cloud module; the three-axis acceleration data acquisition device, the three-axis gyroscope module, and the lidar point cloud module are respectively connected to the data processing module.

[0022] The beneficial effects of this invention are: by calculating the current states of IMU-predicted pose, LiDAR point cloud-matched pose, and RTK observation and localization, and calculating the relative offsets of these three elements, the invention automatically adjusts the fusion method and fusion ratio to automatically create a high-precision map that meets latitude and longitude coordinates. Attached Figure Description

[0023] Figure 1 This is a flowchart illustrating an online high-precision map building method based on location fusion.

[0024] Figure 2 This is a schematic diagram illustrating the principle of an online high-precision map building system based on location fusion.

[0025] Figure 3 A schematic diagram of high-precision mapping by fusing lidar, imu, and RTK. Detailed Implementation

[0026] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings, but the scope of protection of the present invention is not limited to the following description.

[0027] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention; that is, the described embodiments are only a part of the embodiments of the invention, and not all of them. The components of the embodiments of the invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0028] Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention. It should be noted that relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations.

[0029] Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0030] The features and performance of the present invention will be further described in detail below with reference to embodiments.

[0031] like Figure 1 As shown, an online high-precision map building method based on location fusion includes the following steps:

[0032] The three-axis acceleration data and three-axis gyroscope data are acquired through the IMU. The changes in velocity and angle are obtained from the three-axis acceleration data and three-axis gyroscope data, respectively. Based on the changes in velocity and angle, the predicted continuous attitude change data is obtained.

[0033] The local map is built by matching the lidar point cloud and the driving trajectory is obtained. The driving trajectory is matched with the local map to obtain lidar observation pose data.

[0034] Obtain satellite positioning trajectory data fed back by RTK;

[0035] By fitting the predicted continuous attitude change data, lidar observation pose data, and satellite positioning trajectory data at the same moment, the degree of overlap in a unified coordinate system is obtained, and the variance of the relative deviation is obtained.

[0036] The positioning accuracy of the predicted continuous attitude change data, lidar observed pose data, and satellite positioning trajectory data at the same moment is judged by the variance value of the obtained relative deviation, so as to obtain the pose of the sub-map at the same moment.

[0037] The method of obtaining the changes in velocity and angle using triaxial acceleration data and triaxial gyroscope data respectively includes:

[0038] The change in velocity is obtained by integrating the triaxial acceleration data, and the change in angle is obtained by integrating the angular velocity obtained from the triaxial gyroscope data. The change in displacement of the vehicle in space is obtained by combining the change in angle and the change in velocity.

[0039] The lidar observation pose data includes three-dimensional spatial coordinates and three-dimensional rotation angle pose data.

[0040] It also includes converting the predicted continuous attitude change data and lidar observed pose data at the same time moment into latitude and longitude coordinates respectively.

[0041] The method of obtaining the degree of overlap in a unified coordinate system through fitting includes:

[0042] By locally optimizing the origin coordinates of the three space curves, the spatial distance error at the same coordinate at each moment is minimized.

[0043] The method for obtaining the variance value of the relative deviation includes:

[0044] By optimizing the fit of three curves, the expected value of each point on the three curves is obtained. Then, the expected value of the square of the deviation between all points on each curve and the expected value of the corresponding point is calculated.

[0045] The method of judging the positioning accuracy of the predicted continuous attitude change data, lidar observed pose data, and satellite positioning trajectory data at the same moment by using the variance value of the obtained relative deviation to obtain the pose of the sub-map at the same moment includes:

[0046] When the RTK and SLAM localizations are consistent, the pose of the submap is determined by finding the minimum deviation between the two. When the RTK and SLAM localizations are inconsistent, the pose of the submap is determined by the constraint relationships between the point cloud and the submap, and between submaps, and the corresponding calculated observation variance matrix.

[0047] An online high-precision map building system based on positioning fusion, which applies a positioning fusion-based online high-precision map building method, includes a data acquisition module, a satellite positioning module, a data processing module, a map building module, and a communication device; the data acquisition module, satellite positioning module, map building module, and communication device are respectively connected to the data processing module.

[0048] The data acquisition module includes a three-axis acceleration data acquisition device, a three-axis gyroscope module, and a lidar point cloud module; the three-axis acceleration data acquisition device, the three-axis gyroscope module, and the lidar point cloud module are respectively connected to the data processing module.

[0049] Specifically, the three-axis acceleration and three-axis gyroscope data are acquired through the IMU, and the predicted continuous attitude change is obtained through integration calculation over a period of time.

[0050] Within the same time period, local maps are built by matching LiDAR point clouds, and the driving trajectory during this time period is obtained, thus acquiring LiDAR observation pose data.

[0051] Finally, the satellite positioning trajectory for the same time period was obtained from the RTK feedback.

[0052] After obtaining the data of the above three entities, the degree of overlap of the three entities in a unified coordinate system is calculated by fitting through appropriate coordinate transformation and compensation, and the variance of the relative deviation of the three entities is also calculated.

[0053] Finally, the relative deviation variance of the three is used to determine the positioning accuracy. When RTK and SLAM positioning are consistent, the pose of the submap during this time period is determined by finding the solution that minimizes the deviation between the two. When RTK and SLAM positioning are inconsistent, the optimal pose of the submap is solved by the constraint relationships between the point cloud and the submap, and between submaps, and the corresponding calculated observation variance matrix.

[0054] The above description is merely a preferred embodiment of the present invention. It should be understood that the present invention is not limited to the forms disclosed herein and should not be construed as excluding other embodiments. It can be used in various other combinations, modifications, and environments, and can be altered within the scope of the concept described herein through the above teachings or related technologies or knowledge. Modifications and variations made by those skilled in the art that do not depart from the spirit and scope of the present invention should be within the protection scope of the appended claims.

Claims

1. A method for online high-precision map building based on location fusion, characterized in that, Includes the following steps: The three-axis acceleration data and three-axis gyroscope data are acquired through the IMU. The changes in velocity and angle are obtained from the three-axis acceleration data and three-axis gyroscope data, respectively. Based on the changes in velocity and angle, the predicted continuous attitude change data is obtained. The local map is built by matching the lidar point cloud and the driving trajectory is obtained. The driving trajectory is matched with the local map to obtain lidar observation pose data. Obtain satellite positioning trajectory data fed back by RTK; By fitting the predicted continuous attitude change data, lidar observation pose data, and satellite positioning trajectory data at the same moment, the degree of overlap in a unified coordinate system is obtained, and the variance of the relative deviation is obtained. The variance of the obtained relative deviation is used to judge the positioning accuracy of the predicted continuous attitude change data, lidar observation pose data and satellite positioning trajectory data at the same time, so as to obtain the pose of the sub-map at the same time. The method of obtaining the degree of overlap in a unified coordinate system through fitting includes: By locally optimizing the origin coordinates of the three space curves, the spatial distance error at the same coordinate at each time point is minimized. The method of judging the positioning accuracy of the predicted continuous attitude change data, lidar observed pose data, and satellite positioning trajectory data at the same moment by using the variance value of the obtained relative deviation to obtain the pose of the sub-map at the same moment includes: When the RTK and SLAM localizations are consistent, the pose of the submap is determined by finding the minimum deviation between the two. When the RTK and SLAM localizations are inconsistent, the pose of the submap is determined by the constraint relationships between the point cloud and the submap, and between submaps, and the corresponding calculated observation variance matrix.

2. The online high-precision map building method based on location fusion according to claim 1, characterized in that, The method of obtaining the changes in velocity and angle using triaxial acceleration data and triaxial gyroscope data respectively includes: The change in velocity is obtained by integrating the triaxial acceleration data, and the change in angle is obtained by integrating the angular velocity obtained from the triaxial gyroscope data. The change in displacement of the vehicle in space is obtained by combining the change in angle and the change in velocity.

3. The online high-precision map building method based on location fusion according to claim 1, characterized in that, The lidar observation pose data includes three-dimensional spatial coordinates and three-dimensional rotation angle pose data.

4. The online high-precision map building method based on location fusion according to claim 1, characterized in that, It also includes converting the predicted continuous attitude change data and lidar observed pose data at the same time moment into latitude and longitude coordinates respectively.

5. The online high-precision map building method based on location fusion according to claim 1, characterized in that, The method for obtaining the variance value of the relative deviation includes: By optimizing the fit of three curves, the expected value of each point on the three curves is obtained. Then, the expected value of the square of the deviation between all points on each curve and the expected value of the corresponding point is calculated.

6. An online high-precision map building system based on location fusion, applying the location fusion-based online high-precision map building method according to any one of claims 1-5, characterized in that, It includes a data acquisition module, a satellite positioning module, a data processing module, a map building module, and a communication device; the data acquisition module, satellite positioning module, map building module, and communication device are respectively connected to the data processing module.

7. The online high-precision map building system based on positioning fusion according to claim 6, characterized in that, The data acquisition module includes a three-axis acceleration data acquisition device, a three-axis gyroscope module, and a lidar point cloud module; the three-axis acceleration data acquisition device, the three-axis gyroscope module, and the lidar point cloud module are respectively connected to the data processing module.

Images