Autonomous Vehicle Fault Injection for Whole-Vehicle Response Testing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing simulation techniques for vehicle systems are limited to verifying the operation of individual components and systems, making it difficult to determine whether the entire vehicle is operating properly, especially in autonomous vehicles where safety is a critical concern.
Innovation Solution
Simulating faults in vehicle components by injecting faults into one or more components or systems, allowing the entire vehicle response to be determined, including how all systems work together to bring the vehicle to a safe state, using a computing device to control control lines and dynamometers to measure vehicle dynamics.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If existing simulation techniques are used to verify individual vehicle components, then component-level functionality can be validated, but the ability to verify entire vehicle operation is insufficient
Solution Approach 1:
The patent combines multiple individual component simulations into a unified vehicle-level simulation system. The HIL testbed integrates simulations of the autonomous vehicle, fault injection module, and dynamometer into a single coordinated system that can verify entire vehicle operation rather than isolated components.
Solution Approach 2:
The simulation system is designed to perform multiple functions: it can simulate individual component failures, inject faults into various vehicle systems, measure vehicle responses through the dynamometer, and validate overall vehicle safety protocols. This multi-functional capability allows the same system to verify both component-level and vehicle-level operation.
2Reliability
If real-world fault testing is performed on autonomous vehicles, then actual vehicle response to failures can be observed, but safety risks and operational limitations increase
Solution Approach 1:
The patent creates a virtual copy of the autonomous vehicle within the HIL simulation environment. The simulated vehicle replicates the behavior and response characteristics of the physical vehicle, allowing faults to be injected and tested in the virtual model without exposing the actual vehicle to safety risks. The dynamometer provides physical measurement feedback to validate the simulation accuracy.
Solution Approach 2:
The system prepares for potential testing issues by using the dynamometer to physically restrain and measure the vehicle during simulation. This provides a safety buffer that prevents uncontrolled vehicle movements even if software faults cause unexpected behavior, allowing aggressive fault injection without real-world safety concerns.
3Reliability
If comprehensive vehicle system testing is implemented, then overall vehicle safety can be verified, but system complexity and testing resources increase
Solution Approach 1:
The patent divides the complex vehicle system into manageable simulation modules that can be independently configured and tested. The HIL testbed allows selective activation of different vehicle systems and fault scenarios, breaking down comprehensive testing into discrete, controllable segments rather than requiring all systems to be tested simultaneously.
Solution Approach 2:
The dynamometer serves as an intermediary between the virtual simulation environment and physical measurement. It translates virtual vehicle commands into physical responses that can be measured and fed back into the simulation, simplifying the interface between software simulation and physical verification without requiring direct complex integration of all vehicle systems.
Data Source
AI summary
Techniques for fault injection testing are described herein. The techniques may include receiving, at a computing device, an indication to simulate a fault associated with a component of a vehicle. The computing device may be coupled to the component via a control line that includes a relay or switch component that, when activated, causes the component to simulate the fault. The computing device may also receive data indicative of a vehicle response, such as a measured trajectory of the vehicle in response to the fault, whether a backup system of the vehicle performed correctly in response to the fault, whether the vehicle responded to the fault within a threshold period of time, and the like. The computing device may determine a difference between the vehicle response and an intended response of the vehicle relative to a threshold difference.