Trajectory Planner Testing with Rule-Graph Safety Assessment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for testing autonomous vehicle trajectory planners require vast amounts of driving data and lack flexibility in adapting to different safety models and scenarios, making it challenging to ensure the high safety standards needed for autonomous driving systems.
Innovation Solution
A computer system with a test oracle that applies a rule graph comprising extractor and assessor nodes to evaluate the performance of a trajectory planner in real or simulated scenarios, allowing for the creation of custom rules and visualization of results through a graphical user interface, enabling efficient and flexible testing across various safety models and scenarios.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If vast quantities of driving data are collected to demonstrate required safety level, then safety assurance is improved, but data collection time and resource requirements increase significantly
Solution Approach 1:
The patent creates virtual copies of driving scenarios through simulation environments. Instead of collecting vast quantities of real driving data, the system generates synthetic scenario data that replicates real-world driving conditions, allowing safety testing without actual road deployment. This copying approach dramatically reduces data collection time while maintaining safety validation effectiveness.
Solution Approach 2:
The patent performs preliminary safety validation through simulation before actual deployment. By pre-evaluating trajectory planners in virtual environments using rule-based safety models, the system identifies and corrects safety issues beforehand, reducing the need for extensive post-deployment data collection and accelerating the overall safety assurance process.
2Loss of time
If rule-based safety models are used to reduce data requirements, then data collection burden is reduced, but flexibility in adapting to different safety models decreases
Solution Approach 1:
The patent implements a dynamic rule graph structure that can be reconfigured to adapt to different safety models. The system allows rules to be added, removed, or modified based on the specific safety model being evaluated, enabling flexible adaptation while maintaining the efficiency of simulation-based testing. This dynamic configuration capability resolves the contradiction between reduced data collection and model adaptability.
Solution Approach 2:
The patent creates a universal testing platform that can evaluate multiple different safety models through a common simulation infrastructure. The rule-based safety models are designed to be interchangeable within the same framework, allowing the system to maintain reduced data collection requirements while accommodating various safety assessment approaches through a single multi-functional platform.
3Reliability
If extensive safety testing is performed to ensure high safety standards, then safety assurance is improved, but testing complexity and resource requirements increase
Solution Approach 1:
The patent segments the safety testing process into distinct modular components: scenario generation, trajectory planning, rule-based safety evaluation, and result analysis. Each component is independently implemented and can be selectively configured based on testing needs. This segmentation reduces overall system complexity while enabling comprehensive safety testing through coordinated operation of specialized modules.
Data Source
AI summary
A computer system receives scenario data generated using a trajectory planner to control an ego agent responsive to at least one other agent in a real or simulated scenario. A test oracle provides predetermined extractor functions for extracting time-varying numerical signals from the scenario data and predetermined assessor functions for assessing the extracted time-varying signals. The test oracle applies, to the scenario data, a rule graph comprising extractor nodes and assessor nodes. Each extractor node applies one of the predetermined extractor functions to the scenario data to extract an output in the form of a time-varying numerical signal. Each assessor node has one or more child nodes, each child node being one of the extractor nodes or another of the assessor nodes, and the assessor node applies one of the predetermined assessor functions to the output(s) of its child node(s). The test oracle provides an output graph comprising the output of at least one of the assessor nodes and the output(s) of at least one of its child node(s).


