Autonomous Vehicle Positioning Using Traffic Participant Reference Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Self-driving vehicles face driving safety issues when GPS signals are weak, as they cannot update positioning information in time, leading to potential safety hazards.
Innovation Solution
A method and apparatus that obtain reference data, including state and running path information of traffic participants within a preset distance, to update the self-driving vehicle's positioning information, using detection devices like cameras, radar, and infrared devices, and control the vehicle to follow a preset driving path based on this data, even in areas with weak GPS signals.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the self-driving vehicle relies on GPS to update positioning information, then the positioning can be updated globally, but the positioning cannot be updated in time when GPS signal is weak
Solution Approach 1:
The patent introduces traffic participants (other vehicles, infrastructure) as intermediaries to establish a distributed positioning network. When GPS signals are weak, the self-driving vehicle uses detection devices to obtain reference data from these intermediaries, including their positioning information and running paths, to calculate its own position through relative positioning methods, thus resolving the contradiction between global positioning coverage and real-time positioning reliability.
Solution Approach 2:
The patent combines multiple positioning methods (GPS positioning, inertial navigation, and relative positioning based on traffic participant reference data) into a fused positioning system. This merging allows the vehicle to switch between different positioning modes depending on signal conditions, ensuring continuous and timely positioning updates even when GPS signals are weak or unavailable.
2Reliability
If the self-driving vehicle uses detection devices to obtain reference data from traffic participants, then positioning can be updated in weak GPS areas, but the device complexity increases
Solution Approach 1:
The patent makes existing detection devices (cameras, radar, lidar) perform multiple functions: they not only detect traffic participants for safety purposes but also capture reference data for positioning. This multi-functionality allows the system to achieve reliable positioning in weak GPS areas without adding dedicated positioning hardware, thus resolving the contradiction between positioning reliability and device complexity.
Solution Approach 2:
The patent enables traffic participants to provide their own positioning information and running path data as reference data, making the system self-sustaining. Each vehicle and infrastructure component serves as both a user and a provider of positioning reference data, eliminating the need for complex centralized positioning infrastructure and reducing overall system complexity while maintaining high positioning reliability.
Data Source
AI summary
The present application provides a self-driving vehicle positioning method, an apparatus and a storage medium, where the method includes: obtaining reference data, where the reference data includes: state information and running path information of at least one traffic participant within a preset distance from a self-driving vehicle, and updating positioning information of the self-driving vehicle according to the reference data. In the technical solutions, when a GPS signal is weak, the self-driving vehicle can also update the positioning information in time, thereby ensuring driving safety of the self-driving vehicle, and solving a driving safety problem existing in the self-driving vehicle of the prior art.