CN-121977561-A - Fusion positioning method and device in open scene, electronic equipment and storage medium

CN121977561ACN 121977561 ACN121977561 ACN 121977561ACN-121977561-A

Abstract

The application relates to a fusion positioning method, a fusion positioning device, electronic equipment and a storage medium in an open scene, which are applied to the technical field of automatic driving, wherein the method comprises the steps of detecting whether a vehicle driving area is an open area or a non-open area based on point cloud data; the method comprises the steps of detecting that a vehicle enters an open area from a non-open area, switching a positioning mode to a positioning mode based on integrated navigation, processing collected integrated navigation data by using the positioning mode to obtain an integrated navigation positioning result, compensating the integrated navigation positioning result in real time according to the positioning deviation of the pre-counted positioning mode based on point cloud map matching and the integrated navigation positioning mode, taking the compensated positioning result as a final positioning result, matching the compensated positioning result with the point cloud map if the vehicle enters the non-open area from the open area, and switching the positioning mode to the positioning mode based on the point cloud map matching if the matching is successful. Stable high-precision positioning can be realized.

Inventors

SONG KAI
DOU FENGQIAN
BAI YUNLONG

Assignees

云创智行科技(湖州)有限公司

Dates

Publication Date: 20260505
Application Date: 20260121

Claims (10)

1. The fusion positioning method under the open scene is characterized by comprising the following steps: collecting point cloud data around a vehicle, and detecting whether a vehicle running area is an open area or a non-open area based on the point cloud data; If the vehicle is detected to enter an open area from a non-open area, switching a positioning mode from a positioning mode based on point cloud map matching to a positioning mode based on combined navigation; Processing the collected integrated navigation data by utilizing a positioning mode based on integrated navigation to obtain an integrated navigation positioning result; Compensating the combined navigation positioning result in real time according to the positioning deviation of the pre-counted positioning mode based on the point cloud map matching and the positioning mode based on the combined navigation, and taking the compensated positioning result as a final positioning result; If the fact that the vehicle enters the non-open area from the open area is detected, the compensated positioning result is matched with the point cloud map, and if the matching is successful, the positioning mode is switched from the positioning mode based on the integrated navigation to the positioning mode based on the point cloud map matching.
2. The method according to claim 1, wherein the method further comprises: Continuously monitoring the motion state data of the vehicle in the fusion positioning process; when the motion state data is abnormal, corresponding abnormal information is reported, so that the vehicle can determine whether to enter a safety mode or not based on the abnormal information.
3. The method of claim 1, wherein the method of calculating the positioning bias comprises: Processing the point cloud data of the non-open areas at a plurality of historical moments by using a positioning mode based on the point cloud map matching, and determining the point cloud map matching positioning results of the non-open areas at the plurality of historical moments; processing the integrated navigation data of the non-open areas at the plurality of historical moments by utilizing a positioning mode based on integrated navigation, and determining an integrated navigation positioning result of the non-open areas at the plurality of historical moments; And counting the average value of the differences between the point cloud map matching positioning results and the combined navigation positioning results of the non-open areas at the plurality of historical moments, and taking the average value as positioning deviation.
4. The method of claim 1, wherein detecting that the area in which the vehicle is traveling is an open area or a non-open area based on the point cloud data comprises: Sparsity analysis is carried out on the single-frame point cloud data to obtain corresponding point cloud density; If the point cloud density is smaller than or equal to a preset threshold value, determining that the area where the vehicle runs is an open area; and if the point cloud density is greater than a preset threshold value, determining that the area where the vehicle runs is a non-open area.
5. A fusion positioning device in an open scene, the device comprising: the point cloud data acquisition module is used for acquiring point cloud data around the vehicle; the driving area detection module is used for detecting whether an area where the vehicle is driving is an open area or a non-open area based on the point cloud data; The first switching module is used for switching the positioning mode from the positioning mode based on the point cloud map matching to the positioning mode based on the combined navigation if the vehicle is detected to enter the open area from the non-open area; the integrated navigation positioning result determining module is used for processing the acquired integrated navigation data by utilizing a positioning mode based on integrated navigation to obtain an integrated navigation positioning result; The positioning result compensation module is used for compensating the integrated navigation positioning result in real time according to the positioning deviation of the pre-counted positioning mode based on the point cloud map matching and the positioning mode based on the integrated navigation, and taking the compensated positioning result as a final positioning result; The point cloud map matching module is used for matching the compensated positioning result with the point cloud map if the fact that the vehicle enters the non-open area from the open area is detected; And the second switching module is used for switching the positioning mode from the positioning mode based on the integrated navigation to the positioning mode based on the point cloud map matching if the matching is successful.
6. The apparatus of claim 5, wherein the apparatus further comprises: the motion state data monitoring module is used for continuously monitoring the motion state data of the vehicle in the fusion positioning process; And the abnormal information reporting module is used for reporting corresponding abnormal information when the motion state data is abnormal, so that the vehicle can determine whether to enter a safety mode or not based on the abnormal information.
7. The apparatus of claim 5, wherein the apparatus further comprises: The system comprises a positioning deviation determining module, a positioning deviation determining module and a calculating module, wherein the positioning deviation determining module is used for processing point cloud data of a plurality of non-open areas at historical moments by utilizing a positioning mode based on point cloud map matching, determining point cloud map matching positioning results of the non-open areas at the historical moments, processing combined navigation data of the non-open areas at the historical moments by utilizing a positioning mode based on combined navigation, determining combined navigation positioning results of the non-open areas at the historical moments, counting average values of differences between the point cloud map matching positioning results of the non-open areas at the historical moments and the combined navigation positioning results, and taking the average values as positioning deviations.
8. The device of claim 5, wherein the driving area detection module is specifically configured to perform sparsity analysis on single-frame point cloud data to obtain a corresponding point cloud density, determine that an area where the vehicle is driving is an open area if the point cloud density is less than or equal to a preset threshold, and determine that the area where the vehicle is driving is a non-open area if the point cloud density is greater than the preset threshold.
9. An electronic device comprising a processor for executing a computer program stored in a memory, which computer program, when executed by the processor, implements the method of any of claims 1-4.
10. A computer readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the method of any of claims 1-4.

Description

Fusion positioning method and device in open scene, electronic equipment and storage medium Technical Field The application relates to the technical field of automatic driving, in particular to a fusion positioning method and device in open scenes, electronic equipment and a storage medium. Background The high-precision positioning of the unmanned vehicle mainly depends on a laser radar (LiDAR) and a pre-constructed high-precision point cloud map, and the accurate positioning of the vehicle under a map coordinate system is realized through a point cloud matching algorithm. The method has centimeter-level precision and stronger robustness in scenes with rich environmental structures such as urban roads, parks and the like and obvious ground feature characteristics. However, in an open scene, because the reflection characteristics in the laser radar scanning range are sparse, the geometric information available for matching in the point cloud is obviously reduced, the point cloud matching algorithm is difficult to converge or easily falls into a local extremum, the positioning result is unstable, even obvious track jump or drift phenomenon occurs, the vehicle track is discontinuous, and the control and perception fusion is affected. Disclosure of Invention In order to solve the technical problems, the application provides a fusion positioning method, a fusion positioning device, electronic equipment, a storage medium and a computer program product in open scenes. According to a first aspect of the present application, there is provided a fusion positioning method in open scene, comprising: collecting point cloud data around a vehicle, and detecting whether a vehicle running area is an open area or a non-open area based on the point cloud data; If the vehicle is detected to enter an open area from a non-open area, switching a positioning mode from a positioning mode based on point cloud map matching to a positioning mode based on combined navigation; Processing the collected integrated navigation data by utilizing a positioning mode based on integrated navigation to obtain an integrated navigation positioning result; Compensating the combined navigation positioning result in real time according to the positioning deviation of the pre-counted positioning mode based on the point cloud map matching and the positioning mode based on the combined navigation, and taking the compensated positioning result as a final positioning result; If the fact that the vehicle enters the non-open area from the open area is detected, the compensated positioning result is matched with the point cloud map, and if the matching is successful, the positioning mode is switched from the positioning mode based on the integrated navigation to the positioning mode based on the point cloud map matching. Optionally, the fusion positioning method under the open scene further includes: Continuously monitoring the motion state data of the vehicle in the fusion positioning process; when the motion state data is abnormal, corresponding abnormal information is reported, so that the vehicle can determine whether to enter a safety mode or not based on the abnormal information. Optionally, the method for calculating the positioning deviation includes: Processing the point cloud data of the non-open areas at a plurality of historical moments by using a positioning mode based on the point cloud map matching, and determining the point cloud map matching positioning results of the non-open areas at the plurality of historical moments; processing the integrated navigation data of the non-open areas at the plurality of historical moments by utilizing a positioning mode based on integrated navigation, and determining an integrated navigation positioning result of the non-open areas at the plurality of historical moments; And counting the average value of the differences between the point cloud map matching positioning results and the combined navigation positioning results of the non-open areas at the plurality of historical moments, and taking the average value as positioning deviation. Optionally, the detecting, based on the point cloud data, that the area where the vehicle is traveling is an open area or a non-open area includes: Sparsity analysis is carried out on the single-frame point cloud data to obtain corresponding point cloud density; If the point cloud density is smaller than or equal to a preset threshold value, determining that the area where the vehicle runs is an open area; and if the point cloud density is greater than a preset threshold value, determining that the area where the vehicle runs is a non-open area. According to a second aspect of the present application, there is provided a fusion positioning device in open scene, comprising: the point cloud data acquisition module is used for acquiring point cloud data around the vehicle; the driving area detection module is used for detecting whether an area where the vehicle is driving is an