Adaptive Steering Control for Autonomous Vehicle Trajectory Tracking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current autonomous driving systems for terrestrial vehicles face challenges in effectively tracking a predetermined trajectory, particularly in lateral control, as they do not account for the future evolution of the reference trajectory, leading to delays and inefficiencies in maintaining the vehicle's position and direction.
Innovation Solution
The system combines lateral offset control and look-ahead error control, using the knowledge of the reference trajectory's evolution over time to minimize both the lateral offset and the derivative of the look-ahead error, allowing for adaptive control of the vehicle's steering angle to anticipate and maintain the trajectory.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If classic lateral control methods are used that only consider current position and direction, then the control system is simple to implement, but the vehicle cannot anticipate future trajectory changes leading to tracking delays and errors
Solution Approach 1:
The control system calculates the derivative of the look-ahead error with respect to time, which represents the rate of change of the lateral offset. By using this derivative information, the system anticipates future trajectory changes before they occur, allowing proactive adjustment of steering angle rather than reactive correction, thus improving tracking precision without requiring complex additional hardware
Solution Approach 2:
The system implements a feedback mechanism that continuously monitors both the lateral offset (look-ahead error) and its rate of change (derivative). This dual-parameter feedback allows the controller to adjust the steering angle based on both current position error and the speed at which the error is changing, improving tracking accuracy while maintaining a manageable control architecture
2Speed
If the system uses only current lateral offset information for control, then the response is immediate and simple, but delays occur when tracking curved trajectories
Solution Approach 1:
By calculating the time derivative of the look-ahead error, the system obtains information about the rate of change of lateral offset. This derivative represents predictive information about future trajectory deviations, allowing the controller to prepare corrective steering actions in advance, thereby reducing the effective tracking delay without sacrificing immediate responsiveness
Solution Approach 2:
The control system dynamically adapts by using the derivative of the look-ahead error, which changes continuously as the vehicle progresses along the trajectory. This dynamic approach allows the system to respond differently to varying curvature conditions, maintaining optimal response speed while minimizing tracking delays through adaptive steering control
Data Source
AI summary
A method for controlling the lateral movement of a terrestrial vehicle arranged to track a predetermined trajectory, particularly in an assisted driving or autonomous driving scenario, comprising: determining a lateral offset of the vehicle center of mass from the predetermined trajectory; determining a look-ahead error defined as a distance of a virtual look-ahead position of the vehicle center of mass from the predetermined trajectory; and controlling the steering angle of the vehicle so as to also minimize the lateral offset and the first derivative of said look-ahead error over time.