Vehicle Pose Localization Using Map-Based Lane Offset Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current localization techniques for autonomous driving, such as GPS and sensor fusion, face significant accuracy challenges in urban environments due to systematic errors and interference, leading to inadequate lane-level positioning, which is crucial for safe and stable autonomous navigation.
Innovation Solution
The method employs an optimal trajectory estimation using the expectation-maximization (EM) framework to iteratively search for the best lateral and heading offsets, minimizing errors and ensuring convergence, thereby providing accurate vehicle pose localization on a digital map.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If GPS or similar sensors are used for localization, then positioning can be provided, but accuracy is insufficient at lane level due to systematic errors and urban environment interference
Solution Approach 1:
The patent introduces map data as an intermediary reference system between the vehicle and the physical environment. By comparing sensor measurements against pre-stored map features (lane geometries, road boundaries), the system achieves lane-level localization accuracy despite GPS errors. The map acts as a mediator that transforms coarse GPS positions into precise lane-level locations through matching algorithms.
Solution Approach 2:
The system implements iterative feedback loops where localization estimates are continuously refined. The EM algorithm performs repeated cycles of expectation (comparing predicted sensor readings with actual measurements) and maximization (updating pose estimates), with each iteration reducing localization error. This feedback mechanism progressively improves accuracy from initial coarse GPS positions to final lane-level precision.
2Measurement precision
If map matching is performed to improve localization accuracy, then lane-level positioning is achieved, but computational complexity increases
Solution Approach 1:
The patent segments the localization problem into distinct computational stages: (1) initial pose estimation from GPS coordinates, (2) map feature identification and filtering, (3) EM algorithm iterations for refinement, and (4) final pose determination. By dividing the complex map matching task into manageable segments with specific functions, the system reduces overall computational complexity while maintaining lane-level accuracy.
Solution Approach 2:
The system performs preliminary actions by pre-processing and storing map data in optimized formats before runtime. Map features are pre-identified, categorized, and indexed during map creation, allowing rapid retrieval and matching during localization. This preliminary preparation significantly reduces real-time computational requirements when performing actual vehicle localization.
Data Source
AI summary
An approach is provided for localizing a vehicle pose on a map. The approach involves, receiving an input specifying the vehicle pose with respect to a road lane of the map. The approach also involves searching over a set of candidate lateral offsets to select a lateral offset that minimizes a lateral error between the vehicle position with the lateral offset applied and a lateral location of the road lane, wherein the lateral location and the travel direction of the lane are determined from the map. The approach further involves searching over a set of candidate vehicle headings at the selected lateral offset to select a vehicle heading that minimizes a heading error. The approach further involves determining a local optimum of the vehicle pose based on the selected lateral offset and vehicle heading, wherein the vehicle pose is localized to the map based on the local optimum.