Automotive Radar Cross-Path Detection Using Dynamic Alert Zones
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional automotive radar systems for cross-path detection often result in inaccurate target object detection due to the use of predetermined alert zones that are either over- or under-inclusive for varying cross-path angles, leading to false or delayed collision indications.
Innovation Solution
The system employs an automotive proximity sensing detector with a processor that tracks the relative position of a target object over multiple time points, determining a cross-path angle by solving a multiple hypothesis problem to dynamically adjust the alert zone based on the estimated angle, thereby optimizing the region of interest for cross-path detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a predetermined alert zone is used for cross-path detection, then the detection coverage is ensured, but the detection accuracy deteriorates due to over- or under-inclusive zones for varying cross-path angles
Solution Approach 1:
The alert zone is transformed from a static predetermined region to a dynamic region that adapts to the actual cross-path angle. The system continuously updates the alert zone boundaries based on real-time tracking of target object trajectories and calculation of the cross-path angle between the host vehicle heading and target object trajectory, ensuring the zone remains optimally sized for each specific detection scenario
Solution Approach 2:
The system changes the geometric parameters of the alert zone (boundaries, size, orientation) based on the calculated cross-path angle. By adjusting the zone parameters dynamically according to the actual angle rather than using fixed predetermined values, the system achieves both comprehensive coverage and high detection accuracy for varying cross-path scenarios
2Device complexity
If a predetermined alert zone is used for cross-path detection, then the system complexity is reduced, but the detection accuracy deteriorates due to false or delayed collision indications
Solution Approach 1:
The system replaces complex mechanical or geometric pre-programming of multiple fixed alert zones with a computational approach. Instead of implementing multiple predetermined zones for different angles, the system uses processors to calculate the cross-path angle and dynamically generate the appropriate alert zone, trading mechanical/geometric complexity for algorithmic simplicity
Solution Approach 2:
A single dynamic alert zone generation mechanism serves multiple cross-path angle scenarios. The same processor-based system can handle 45-degree angles, 90-degree angles, and any intermediate angle by simply adjusting the zone parameters, eliminating the need for multiple specialized predetermined zones
3Reliability
If overlapping regions of interest are selected to represent an aggregate alert zone, then the coverage for multiple cross-path scenarios is improved, but the false indication of impending collision increases
Solution Approach 1:
Instead of using a uniform aggregate zone that treats all areas equally, the system applies local quality by adjusting the alert zone boundaries based on the specific cross-path angle. Each region of the alert zone is optimized for its particular angular scenario, providing appropriate sensitivity locally rather than applying a blanket coverage approach that causes false indications
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach enhances the accuracy of cross-path detection by adaptively adjusting the alert zone according to the cross-path angle, reducing false alarms and improving the timeliness of collision warnings, and can be applied to both stationary and moving host vehicles.
Implementation Method 1
an automotive proximity sensing detector including a radar transmitter for transmitting radar signals toward the target object; and a radar receiver for receiving a plurality of returning radar signals reflected from the target object
Data Source
AI summary
Systems and methods are presented herein for improved cross-path detection between a host vehicle and a target object. In general, the cross-path angle by solving a multiple hypothesis problem characterized by independent calculations across a first plurality of time points, the multiple hypothesis problem supposing a plurality of possible cross-path angle solutions, each cross-path angle solution representing a corresponding possible trajectory for the target object. Advantageously, a cross-traffic alert or other feedback may be triggered based at least partially on whether the target object is within a region of interest, wherein the region of interest is determined at least in part based on the estimated cross-path angle. In some embodiments the cross-path angle may be determined by selecting a cross-path angle solution from the plurality of possible cross-path angle solutions which minimizes a variance between results of the independent calculations across the first plurality of time points. Other feature disclosed herein include, determining an estimate for the cross-path angle for each of a plurality of trackings of the target object, wherein each tracking is characterized by a different set of two or more time points. This may advantageously enable, determining a change in trajectory of the target object based on variance between cross-path angle estimates.