Autonomous Vehicle Trajectory Validation Using a Trusted Planner
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicle planning systems require rigorous testing for safety and reliability, especially when changes occur, but extensive real-world testing is costly and time-consuming.
Innovation Solution
A method where a first trajectory generated by an 'untrusted' planning system is validated by a 'trusted' planning system by inputting a portion of the first trajectory into the trusted system, generating a second trajectory, and determining its safety based on various constraints before enabling the autonomous vehicle to follow the first trajectory.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If extensive real-world testing is conducted to validate planning systems, then safety and reliability are improved, but cost and time consumption increase significantly
Solution Approach 1:
The patent applies preliminary action by performing validation checks before actual deployment. A validation planning system generates validation trajectories in advance to verify safety constraints before the autonomous vehicle operates with the testing planning system, preventing unsafe operations from occurring in the first place
Solution Approach 2:
The patent uses an intermediary approach by introducing a validation planning system as a mediator between the testing planning system and the actual vehicle operation. This intermediary system generates validation trajectories that check safety constraints without requiring extensive real-world testing of the entire system
2Reliability
If extensive real-world testing is conducted to validate planning systems, then safety and reliability are improved, but cost increases significantly
Solution Approach 1:
The validation planning system serves as an intermediary that performs safety verification through computational trajectory generation rather than expensive real-world testing. This intermediary layer filters out unsafe trajectories before they reach the vehicle, reducing the need for costly physical test runs
Solution Approach 2:
The patent uses copying by creating virtual validation trajectories through the validation planning system that replicate the safety-checking function without requiring physical vehicle operation. These computational copies allow extensive testing to be performed in silico rather than in the expensive real world
3Productivity
If a new planning system is deployed without rigorous testing, then productivity is improved, but safety and reliability deteriorate
Solution Approach 1:
The system performs preliminary validation trajectory generation before deployment of new planning systems. The validation planning system creates safety-check trajectories in advance, allowing rapid deployment of new planners while maintaining safety through pre-computed validation checks
Solution Approach 2:
The validation process operates continuously alongside the testing planning system without interrupting the deployment workflow. The validation planning system continuously generates validation trajectories to verify safety constraints, enabling ongoing productivity while maintaining constant safety oversight
Data Source
AI summary
Aspects of the disclosure provide controlling an autonomous vehicle. For example, a first trajectory generated by a first planning system may be received. A portion of the first trajectory may be input into a second planning system. A second trajectory generated by the second planning system based on the portion of the first trajectory may be received. The second trajectory may be used to determine whether to validate the first trajectory, and when the second trajectory is determined to be validated, the autonomous vehicle may be enabled to be controlled based on the first trajectory.