Radar Lane Allocation Using Roadway Edge References
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for lane allocation of dynamic objects in radar measurements are prone to errors due to inaccuracies in vehicle orientation, leading to incorrect lane assignments, especially for distant objects.
Innovation Solution
A method that subdivides detected objects into dynamic and static, determines the roadway edge from static objects, and uses heuristic methods to allocate lanes based on the distance of dynamic objects to these edges, eliminating orientation-dependent errors.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If vehicle orientation and map-based localization are used for lane allocation, then lane assignment can be performed, but measurement precision deteriorates due to orientation errors especially for distant objects
Solution Approach 1:
The patent extracts the lane allocation function from the map-based localization system and implements it independently using only radar measurements. By separating the lane allocation task from the orientation-dependent localization process, the system eliminates the propagation of orientation errors to lane assignment, achieving reliable lane allocation without requiring accurate vehicle orientation or map data.
Solution Approach 2:
The patent introduces static objects (roadside infrastructure) as intermediaries to establish a reference frame for lane allocation. These static objects serve as mediators between the radar measurements and lane assignment, providing stable reference points that are independent of vehicle orientation. The dynamic objects' positions are evaluated relative to these static references rather than relative to the moving vehicle's orientation.
2Ease of operation
If map-based localization with global position and pose is used, then lane allocation can be realized, but device complexity increases
Solution Approach 1:
The patent extracts the essential function of lane allocation from the complex map-based localization system. By taking out only the necessary radar measurement data and static object detection capability, the system achieves lane allocation without requiring full map matching, global position determination, or pose estimation, thereby significantly reducing device complexity.
Solution Approach 2:
The patent replaces the expensive and complex map data structure with simple, transient radar measurements of static objects. Instead of maintaining and processing detailed map information, the system uses temporary detections of roadside infrastructure from radar, which are processed only when needed for lane allocation, reducing both computational and data storage requirements.
3Device complexity
If radar measurements are used directly for lane allocation without localization, then device complexity is reduced, but measurement precision may be affected
Solution Approach 1:
The patent uses static objects detected by radar as intermediaries to establish a stable reference frame. These static roadside objects mediate between the raw radar measurements and the final lane allocation decision, providing consistent reference points that enable accurate lane assignment using only simple radar data without requiring complex localization systems.
Data Source
AI summary
A method for allocating dynamic objects to traffic lanes of a road. Objects are detected in a radar measurement and are subdivided into dynamic objects and static objects. A roadway edge is determined from the static objects. Here, the lane allocation of the dynamic object is made from the distance of a dynamic object to at least one static object marking the roadway edge.