Autonomous Vehicle Localization Using Predefined Environmental Features
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles face challenges in determining their location when GPS and LiDAR data is unavailable or inaccurate, such as in tunnels or bridges, where traditional localization methods fail to provide reliable accuracy.
Innovation Solution
The system switches to a second localization mode that uses predefined features of the environment, such as textural patterns on the road surface or installed markers, to determine the vehicle's location when traditional data sources do not meet an accuracy criterion.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If GPS and LiDAR localization methods are used, then location determination works in open environments, but localization accuracy deteriorates in tunnels or on bridges where signals are unavailable
Solution Approach 1:
The system dynamically switches between two localization modes based on environmental conditions. In the first mode, GPS and LiDAR are used for location determination. When these methods fail to meet accuracy criteria (such as in tunnels or on bridges), the system transitions to the second mode using predefined environmental features. This dynamic adaptation ensures reliable localization across diverse environments.
Solution Approach 2:
The system changes the localization parameters and data sources based on environmental context. Instead of relying solely on GPS coordinates and LiDAR point clouds, the system switches to using predefined features such as road surface textural patterns, curb characteristics, and other environmental markers when traditional methods become unreliable, thereby maintaining measurement precision.
2Ease of manufacture
If traditional localization methods (GPS, LiDAR) are used, then the system is simple to implement, but localization fails in environments with signal unavailability
Solution Approach 1:
The localization system is designed with multi-functionality to handle both open environments and signal-denied environments. It can operate using GPS and LiDAR in normal conditions, and seamlessly switch to using predefined environmental features when signals are unavailable. This universal approach ensures the system remains reliable across all driving environments without requiring completely separate systems.
3Measurement precision
If the system switches to predefined features for localization, then accuracy is maintained in signal-denied areas, but system complexity increases
Solution Approach 1:
Predefined environmental features are identified, mapped, and stored in advance during system setup or map creation phases. These features include road surface textural patterns, curb geometries, and other distinctive environmental markers with known locations. When the vehicle enters signal-denied areas, the system can immediately utilize these pre-prepared features for localization without requiring complex real-time feature extraction and mapping, thereby limiting the increase in operational system complexity.
Data Source
AI summary
The subject matter described in this specification is directed to a computer system and techniques for determining a location of an autonomous vehicle. The computer system is configured to determine the location using localization data from multiple data sources. When the localization data from the multiple data sources is unavailable or inaccurate, the computer system is configured to determine the location using predefined features of the environment.