Vehicle Trajectory Safety Margins for Controller Delay Compensation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles face performance suboptimality and collision risks due to discrepancies between intended and actual trajectories, caused by delays and dynamics in controllers and actuators, which existing technologies fail to adequately address without real-world operation and direct error measurement.
Innovation Solution
The method involves determining a safety margin by analyzing reference trajectories and error signals using machine-learning models and system identification, allowing for the calculation of safety margins without real-world operation, and adjusting controller gains to minimize these margins, thereby enhancing vehicle safety and efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the vehicle strictly follows the reference trajectory, then collision risk is reduced, but actual trajectory deviation occurs due to controller and actuator delays
Solution Approach 1:
The patent calculates safety margins in advance by analyzing the spectral content of reference trajectories and determining error signals that account for controller and actuator delays. This preliminary calculation allows the vehicle to proactively adjust its trajectory to compensate for expected deviations, rather than reactively responding to collisions or near-misses.
Solution Approach 2:
The patent employs a feedback mechanism where the actual trajectory is continuously compared with the reference trajectory, and the error signal is used to adjust future trajectory calculations. This closed-loop control ensures that trajectory accuracy is maintained despite delays in the controller and actuator system.
2Measurement precision
If extensive real-world testing is conducted to determine safety margins, then measurement precision improves, but time consumption and operational requirements increase
Solution Approach 1:
The patent creates a computational model that replicates the vehicle's dynamics and controller behavior, allowing safety margins to be determined through simulation and analysis of reference trajectory spectra rather than extensive real-world testing. This virtual copying enables accurate safety margin calculation without the time and operational requirements of physical testing.
Solution Approach 2:
The safety margins are calculated in advance using spectral analysis of reference trajectories and system identification methods, before the vehicle operates in real-world conditions. This preliminary determination eliminates the need for time-consuming iterative testing while providing accurate safety margin values for trajectory generation.
3Manufacturing precision
If controller gains are increased to reduce trajectory error, then trajectory accuracy improves, but safety margin increases requiring larger clearance from objects
Solution Approach 1:
The patent optimizes controller gains by analyzing the spectral content of reference trajectories and determining the appropriate gain values that achieve acceptable trajectory accuracy while minimizing safety margins. This parameter optimization ensures that the controller is tuned to the specific characteristics of the reference trajectories it will follow, rather than using conservative fixed gains.
Solution Approach 2:
The patent makes the safety margin dynamic by calculating it based on the spectral characteristics of each reference trajectory and the corresponding error signal. Rather than using a fixed safety margin, the system adapts the margin to the specific trajectory being followed, allowing tighter clearance when the trajectory characteristics and controller performance permit.
Data Source
AI summary
Techniques for determining a safety margin by which to limit trajectory(ies) generated by a vehicle control system such that the vehicle will not exceed the safety margin more than a target occurrence rate. The techniques may include determining a first spectrum associated with trajectory data generated by one or more vehicles, generating a model of a vehicle, and determining a spectrum of an error signal based at least in part on the model and the first spectrum. Determining the safety margin may be based at least in part on the spectrum of the error signal and a target occurrence rate. Operation characteristics of components of the vehicle (e.g., controller, steering actuator) may be tuned based at least in part on the model, first spectrum, and/or second spectrum. The techniques enable determining safety margins for untested vehicles and/or for different operating states of a vehicle.