Road Surface Landmark Localization for High-Resolution Vehicle Positioning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional GPS-based localization systems lack sufficient accuracy and resolution for advanced vehicle features like autonomous driving, and landmark-based methods require costly and time-consuming prior knowledge of landmark locations.
Innovation Solution
A method for identifying landmarks by collecting data during vehicle traversal, dividing it into snippets, calculating distinctiveness scores, and designating distinctive snippets as landmarks, storing their relative locations, and using these landmarks for localization without requiring absolute location knowledge.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If GPS-based localization is used, then the system is simple and widely available, but the localization accuracy and resolution are insufficient for advanced vehicle features
Solution Approach 1:
The patent introduces road surface data as an intermediary element between the vehicle and the localization system. By analyzing road surface characteristics (roughness, texture, patterns) collected by vehicle sensors, the system creates an indirect but highly accurate localization method that doesn't require complex specialized equipment, thus resolving the contradiction between accuracy and complexity
Solution Approach 2:
The patent replaces the traditional GPS satellite-based electromagnetic system with a mechanical sensing approach using vehicle suspension and road contact data. This substitution achieves higher precision localization through mechanical interaction with the road surface rather than relying on satellite signals, addressing the accuracy limitation while maintaining system simplicity
2Measurement precision
If landmark-based localization is used, then the localization accuracy is improved, but it requires costly and time-consuming a priori identification and mapping of landmarks
Solution Approach 1:
The system enables vehicles to automatically identify and map road surface landmarks themselves during normal operation, without requiring external assistance or pre-surveying. The vehicle's existing sensors collect road data, and the processing system automatically extracts distinctive features as landmarks, eliminating the time-consuming manual landmark identification process while maintaining high localization accuracy
Solution Approach 2:
The patent performs preliminary identification of road surface landmarks during the data collection phase, storing their locations in advance. This preliminary action allows the vehicle to quickly reference pre-identified landmarks during localization without needing to perform real-time landmark detection, thus reducing the time loss while maintaining accuracy
3Reliability
If traditional landmark-based localization is used, then localization accuracy is improved, but the system cannot adapt when landmarks are relocated or new routes are traveled
Solution Approach 1:
The patent implements a dynamic landmark database that can be continuously updated with new road surface features as vehicles traverse new routes or as road conditions change. The system adapts to relocated landmarks or new routes by collecting fresh road surface data and updating the landmark repository, ensuring localization reliability is maintained across changing conditions and routes
Solution Approach 2:
The system changes the parameters of the landmark database dynamically, adding new landmarks, removing relocated ones, and updating their characteristics based on current road surface data. This parameter adaptation allows the localization system to maintain reliability when facing route changes or landmark relocation without requiring complete system reconfiguration
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Localization systems, including road surface-based localization systems, are disclosed that may work in conjunction with less precise systems, such as GPS, to accurately locate a vehicle on a road segment. Methods and apparatus are disclosed for the identification and selection of landmarks, including road surface-based landmarks.