HIL Robustness Testing and Validation of Computer Vision and C-ITS for Vehicle Fault Mitigation

As computer vision aided systems are getting more and more common in safety-critical applications, particularly in the automotive sector, ensuring their robustness against both software and hardware faults is of paramount importance. These systems must prevent unintended behaviours, such as frame loss caused by performance bottlenecks, which can lead to hazardous scenarios. Furthermore, the emergence and adoption of Cooperative-Intelligent Transportation Systems have introduced innovative strategies for misbehavior detection and fault mitigation, enabling their evaluation in next-generation scenarios. In this context, we propose a system architecture designed to support computer vision-based vehicular control strategies while facilitating robust testing on a Hardware-in-the-Loop platform. Our approach emphasizes the limits of object detection algorithms under heterogeneous load conditions, crucially due to the absence of tailored real-time protection mechanisms. To address these challenges, we demonstrate the effectiveness of Cooperative Perception as a fault mitigation strategy within this context, highlighting its potential to significantly enhance the driving performance of both human and AI drivers.