Autonomous Vehicle Trajectory Validation Using Trusted Planner Cross-Check
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicle planning systems require rigorous testing due to safety and reliability concerns, but extensive real-world driving is costly and inefficient for validation.
Innovation Solution
A method where a first trajectory generated by an 'untrusted' planning system is validated using a second 'trusted' planning system by inputting a portion of the first trajectory into the second system, generating a second trajectory, and determining its safety to validate the first trajectory.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If extensive real-world driving 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 creates a virtual copy of the real-world driving environment through simulation. The validation system uses simulated trajectories that replicate real driving scenarios, allowing comprehensive safety testing without the time and cost of actual road testing. The simulation environment copies road geometry, traffic conditions, and vehicle dynamics to provide realistic validation.
Solution Approach 2:
The patent performs validation actions in advance through simulated testing before deploying planning systems to real-world operations. The system pre-validates trajectory generators by running extensive simulated miles, identifying safety issues, and certifying systems beforehand, thus avoiding the need for lengthy real-world testing campaigns.
2Reliability
If millions of miles of real-world driving are used for testing, then validation thoroughness is improved, but resource efficiency deteriorates
Solution Approach 1:
The patent replaces physical real-world driving with virtual simulated driving. The validation system processes simulated mileage that can be generated computationally without consuming physical resources like fuel, vehicle wear, or human driver time. This copying approach maintains validation thoroughness while dramatically improving resource efficiency.
Solution Approach 2:
The patent substitutes mechanical real-world testing with computational simulation. Instead of physically driving vehicles for millions of miles, the system uses computer-based trajectory generation and validation algorithms to perform the same safety assessments, replacing mechanical resource consumption with computational processing.
3Productivity
If new planning systems are implemented without rigorous testing, then productivity is improved, but safety risks increase
Solution Approach 1:
The patent performs safety validation in advance through simulated testing before allowing new planning systems to be deployed. The validation system runs comprehensive safety checks on trajectory generators beforehand, certifying them for use. This preliminary validation enables rapid deployment of improved systems while maintaining safety standards.
Solution Approach 2:
The patent introduces a virtual simulation environment as an intermediary between development and real-world deployment. This intermediary validation layer allows new planning systems to be tested thoroughly in a controlled simulated environment before reaching actual road use, filtering out unsafe systems while enabling rapid iteration of safe ones.
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.