Digital Map Alignment Using Radar and Video Boundary Features
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for creating digital maps for automated vehicles face challenges in accurately aligning and integrating radar and video data from multiple vehicles to ensure precise positioning and safe operation, particularly in the lateral direction, which is crucial for safe navigation.
Innovation Solution
A two-step method is employed to create a high-precision digital map by combining radar data values along vehicle trajectories and aligning surrounding area features based on radar signatures, followed by aligning these features relative to boundary features detected by video sensors to enhance positional accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If radar data values from multiple vehicles are combined along their trajectories, then the digital map can be created with comprehensive surrounding area features, but alignment deviations occur particularly in the lateral direction
Solution Approach 1:
Boundary features detected by video sensors serve as an intermediary reference system to mediate between radar data from different vehicles. These boundary features (road edges, curbs, barriers) provide a common spatial reference that enables accurate lateral alignment of radar data from multiple vehicles traveling along the same route.
Solution Approach 2:
The alignment process is segmented into two distinct steps: first aligning radar data along the longitudinal direction based on vehicle trajectories, then separately aligning in the lateral direction using boundary features. This segmentation allows each alignment dimension to be optimized independently with appropriate methods.
2Measurement precision
If a two-step alignment method is used to improve lateral alignment accuracy, then positional precision is enhanced, but the complexity of the map creation process increases
Solution Approach 1:
The complex alignment task is divided into two manageable sub-tasks: longitudinal alignment based on vehicle trajectories and lateral alignment based on boundary features. This segmentation reduces the overall complexity by breaking down the problem into simpler, more tractable components that can be processed sequentially.
Solution Approach 2:
The longitudinal alignment is performed as a preliminary step before lateral alignment. By first establishing the correct longitudinal positioning based on vehicle trajectories, the subsequent lateral alignment using boundary features becomes simpler and more accurate, as the reference frame is already properly established.
3Measurement precision
If radar and video data from multiple vehicles are integrated, then the digital map achieves high precision for automated vehicle navigation, but the data processing requirements and computational load increase
Solution Approach 1:
The data processing is segmented into distinct stages: trajectory-based longitudinal alignment, boundary feature extraction from video data, and boundary feature-based lateral alignment. Each stage processes only the necessary data for its specific purpose, avoiding redundant computations and reducing overall computational load.
Solution Approach 2:
Boundary features are extracted from video data as a separate, simplified representation that serves as a reference for lateral alignment. This extraction process separates the essential alignment information from the full video data, reducing the computational burden while maintaining alignment accuracy.
Data Source
AI summary
A method and first device for creating a digital map. The digital map represents at least one region along a traffic route. A method and second device for operating an automated vehicle along a traffic route is also described.