Point Cloud Key Point Matching for GPS-Denied Vehicle Positioning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing self-driving technologies rely heavily on GPS, which can be unreliable or unavailable in certain environments, leading to challenges in accurate positioning without GPS signals.
Innovation Solution
A positioning method and apparatus that uses point cloud data collected by a to-be-positioned terminal to determine current key points and their distribution features, which are then matched with reference key points in a global positioning map to determine the current pose data of the terminal, allowing for accurate positioning without relying on GPS.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If GPS is used for positioning, then positioning can be achieved in open environments, but positioning becomes unreliable or unavailable in certain environments (e.g., urban canyons, tunnels, dense forests)
Solution Approach 1:
The patent introduces point cloud data as an intermediary medium between the terminal and the positioning system. Instead of directly relying on GPS signals, the system uses point cloud data collected by sensors (lidar, radar, cameras) as an intermediary to establish positioning through feature matching with pre-stored reference point cloud data, thereby achieving reliable positioning in GPS-denied environments
Solution Approach 2:
The patent replaces the traditional GPS-based positioning mechanism with a sensor-based point cloud matching mechanism. By substituting the mechanical/electromagnetic GPS signal system with a computational geometry-based point cloud matching system, the solution achieves positioning functionality in environments where GPS signals are unavailable or unreliable
2Adaptability or versatility
If point cloud data processing is performed locally, then positioning can be achieved without cloud dependency, but computational complexity and data processing requirements increase significantly
Solution Approach 1:
The patent segments the positioning computation into two parts: local processing of point cloud data extraction and feature identification, and remote processing of matching computation. The terminal locally processes point cloud data to extract features and sends only essential information to the cloud server, which performs the computationally intensive matching operation, thereby reducing local computational complexity while maintaining positioning independence
Solution Approach 2:
The cloud server acts as an intermediary that handles the computationally intensive point cloud matching operation. By offloading this complex computation to the cloud, the system reduces the computational burden on the terminal device while maintaining the ability to perform positioning without GPS dependency
Data Source
AI summary
Exemplary positioning method and apparatus, a device, a system, a medium and a self-driving vehicle are provided. The positioning method includes determining, according to the current frame of point cloud data collected by a to-be-positioned terminal in a travelling environment, the current key point in the current frame of point cloud data and a point cloud distribution feature of the current key point; selecting, according to point cloud distribution features associated with reference key points in a global positioning map and the point cloud distribution feature of the current key point, a target key point matching the current key point from the reference key points; and determining, according to reference pose data associated with the target key point, the current pose data of the to-be-positioned terminal.