Vehicle Positioning Using GNSS and Laser Point Cloud Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional vehicle positioning methods face challenges in achieving quick and accurate high-precision positioning, particularly in dynamic environments where environmental changes affect positioning accuracy.
Innovation Solution
A method that combines GNSS positioning with laser point cloud data from easily recognizable road objects, involving data extraction, forward simulation, and matching to determine the vehicle's high-precision position by converting laser point data into a standard coordinate system and calculating probabilities based on matching results.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional GNSS positioning is used, then the positioning system is simple to implement, but the positioning accuracy is limited to meter-level
Solution Approach 1:
The patent combines GNSS positioning with laser point cloud matching to achieve high-precision positioning. The GNSS provides coarse positioning information while the laser point cloud matching refines the position to centimeter-level accuracy, merging two positioning approaches to overcome the limitations of each individual method
Solution Approach 2:
The positioning process is segmented into multiple stages: first obtaining GNSS coarse positioning, then extracting laser point cloud data of road objects, and finally performing point cloud matching to refine the position. This segmentation allows the system to achieve high precision without overwhelming computational complexity at each stage
2Measurement precision
If high-precision map based positioning is used, then centimeter-level positioning accuracy is achieved, but the computational complexity and data processing time increase significantly
Solution Approach 1:
The patent extracts only the essential laser point cloud data of easily recognizable road objects with stable attributes, rather than processing complete high-precision maps. This extraction approach maintains centimeter-level positioning accuracy while significantly reducing data volume and computational complexity, thereby improving positioning speed
3Reliability
If complete laser point cloud data is processed, then comprehensive environmental information is obtained, but the data volume and computational burden increase
Solution Approach 1:
The patent focuses on capturing laser point cloud data of specific road objects with stable attributes (such as road markings, curbs, and permanent structures) rather than processing all environmental objects. This local quality approach ensures reliable positioning by selecting objects that provide consistent reference points while minimizing data volume
Data Source
AI summary
A positioning method, a device, and an electronic apparatus. The method comprises: obtaining a GNSS position of a vehicle; obtaining standard positioning data around a road where the vehicle is located; obtaining laser point cloud data around the road where the vehicle is located; extracting to-be-matched positioning data around the road; forward simulating a motion state of the vehicle based on sampling positions corresponding to the vehicle at a previous moment, to obtain sampling positions at a current moment; converting the to-be-matched positioning data into a coordinate system corresponding to the standard positioning data; matching the to-be-matched positioning data with the standard positioning data to obtain a probability that the vehicle is located at each sampling position at the current moment; and obtaining a position of the vehicle based on the probability that the vehicle is located at each sampling position at the current moment.