Hardware-in-the-Loop-Based Real-Time Fault Injection Framework for Dynamic Behavior Analysis of Automotive Software...

Area of Science:

• Automotive Engineering
• System Safety
• Control Systems

Background:

• Safety-critical systems in vehicles require rigorous reliability and safety assessment.
• Traditional analytical techniques are insufficient for complex automotive systems' dynamic behavior.
• Simulation-based fault injection (FI) offers early-stage fault analysis but needs real-time conditions.

Purpose of the Study:

• To propose a real-time fault injection framework for automotive safety-critical systems.
• To enable experimental analysis of system response under abnormal conditions during development.
• To enhance the accuracy, comprehensiveness, and realism of system behavior analysis.

Main Methods:

• Developed a real-time FI framework using a hardware-in-the-loop (HiL) simulation platform and an electronic control unit (ECU) prototype.
• Modeled the framework in MATLAB/Simulink and implemented it on the HiL system for real-time V-cycle analysis.
• Programmatically injected nine types of sensor and actuator faults (single and multiple) into the HiL system, considering complete vehicle dynamics and environment models.

Main Results:

• Demonstrated the framework's capability to inject transient and permanent faults in real time.
• Analyzed the effects of single and simultaneous faults on system performance.
• Provided experimental results comparing faulty mode performance against a golden running mode.

Conclusions:

• The proposed real-time FI framework effectively analyzes system behavior under fault conditions.
• The hardware-in-the-loop approach facilitates early detection and mitigation of potential safety hazards.
• This method enhances the reliability and safety assessment of automotive electronic control units.