Autonomy Fault Injection for Repeatable AV Subsystem Testing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing safety-testing approaches for autonomous vehicles (AVs) fail to isolate specific subsystems for evaluating responses to different types of faults, leading to incomplete and unreliable assessments of response mechanisms in real-world environments, and require extensive time, human labor, and resources for testing in artificial or controlled environments.
Innovation Solution
A fault-injection subsystem is integrated into the AV's autonomy system to automatically inject predefined faults based on real-world variables, allowing for repeatable testing of response times and actions under various conditions, enabling targeted evaluation of specific subsystems and capturing data for insights into response mechanisms.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If safety testing is performed in artificial or controlled environments, then testing can be conducted with controlled conditions, but extensive time, human labor, and resources are required
Solution Approach 1:
The patent uses simulation environments that create virtual copies of real-world driving scenarios, allowing safety testing to be performed in replicated conditions without requiring physical test tracks or real vehicles for every test case. The simulation copying enables repeated testing of the same scenarios efficiently.
Solution Approach 2:
The system pre-defines multiple fault scenarios and testing protocols before actual safety assessment begins. By preparing test cases, fault injection patterns, and evaluation criteria in advance, the system eliminates the need for extensive on-site setup and human coordination during actual testing.
2Reliability
If fault injection is performed manually in controlled environments, then specific fault scenarios can be tested, but extensive human labor and resources are required
Solution Approach 1:
The safety assessment system performs self-testing by automatically injecting predefined faults into its own autonomy system during simulation runs. The system autonomously monitors its own response to injected faults without requiring external human operators to manually introduce and track each fault scenario.
Solution Approach 2:
The system implements automated feedback loops where fault injection triggers are automatically detected, the autonomy system's response is captured and analyzed, and results are fed back into the safety assessment database. This closed-loop automation eliminates manual intervention in the fault injection and data collection process.
3Adaptability or versatility
If comprehensive safety testing is conducted without subsystem isolation, then all system interactions are captured, but specific subsystem responses cannot be evaluated independently
Solution Approach 1:
The patent segments the autonomy system into distinct functional subsystems (perception, prediction, planning, control) and enables independent fault injection into each subsystem. The testing framework is divided into modular components that can target specific subsystems while maintaining the ability to test integrated system behavior when needed.
Data Source
AI summary
Examples disclosed herein may involve (i) obtaining data for one or more data variables related to autonomous operation of a vehicle in a test environment being facilitated by the vehicle's autonomy system, (ii) based on the obtained data, evaluating one or more predefined fault rules, each of which comprises (a) one or more predefined criteria related to the one or more data variables and (b) a predefined fault that is to be injected into the autonomy system when the one or more predefined criteria are determined to be satisfied, (iii) based on the evaluation, injecting a predefined fault into the autonomy system, and (iv) capturing data indicative of a response by a response mechanism of the vehicle to the vehicle autonomously operating in accordance with the injected fault.