Steering Torque Path Prediction for Faster Collision Intervention
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing advanced driver assistance systems (ADAS) face delays in predicting vehicle trajectories, which can be too late for collision avoidance applications, especially when relying on sensor-based lane information.
Innovation Solution
The method involves computing a predicted steering angle at a future time step using steering signals, determining the corresponding predicted vehicle position, detecting proximity to objects, and generating an intervention signal if the proximity is below a predetermined threshold.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If sensor-based lane information is used for trajectory prediction, then measurement precision is improved, but response time deteriorates
Solution Approach 1:
The system segments the trajectory prediction task into two parts: a fast steering-angle-based prediction for immediate response, and a slower sensor-based lane information for verification and correction. This segmentation allows the system to achieve both speed and accuracy by using the appropriate method for each aspect of the prediction.
Solution Approach 2:
The steering angle serves as an intermediary parameter that bridges the gap between driver input and vehicle position. By computing predicted position directly from steering angle through a vehicle model, the system creates a fast prediction pathway that doesn't depend on sensor processing delays, while sensor data acts as a mediator to correct and refine this prediction.
2Reliability
If sensor-based methods are used for lane detection, then reliability is improved, but processing speed deteriorates
Solution Approach 1:
The system performs preliminary trajectory prediction using the steering angle before sensor processing is complete. This preliminary action provides an immediate prediction that can be used for time-critical interventions, while the sensor-based lane detection runs in parallel to provide reliable verification and correction without blocking the fast response path.
Data Source
AI summary
A steering system includes a motor, and a processor configured to perform a method. The method includes computing, by a processor, a predicted steering angle at a future time step. The method further includes determining, by the processor, a predicted vehicle-position at the future time step based on the predicted steering angle. The method further includes detecting, by the processor, a proximity to an object at the future time step based on predicted vehicle-position. The method further includes generating, by the processor, an intervention signal in response to the proximity being below a predetermined threshold.