Transient Fault Tolerant Semantic Segmentation for Autonomous Driving
This addresses reliability issues in autonomous vehicle perception systems, but it is incremental as it builds on existing hardening techniques.
The paper tackled the problem of hardware fault-tolerance in semantic segmentation models for autonomous driving by introducing ReLUMax, a novel activation function that improves robustness without time overhead, preserving performance and boosting prediction confidence.
Deep learning models are crucial for autonomous vehicle perception, but their reliability is challenged by algorithmic limitations and hardware faults. We address the latter by examining fault-tolerance in semantic segmentation models. Using established hardware fault models, we evaluate existing hardening techniques both in terms of accuracy and uncertainty and introduce ReLUMax, a novel simple activation function designed to enhance resilience against transient faults. ReLUMax integrates seamlessly into existing architectures without time overhead. Our experiments demonstrate that ReLUMax effectively improves robustness, preserving performance and boosting prediction confidence, thus contributing to the development of reliable autonomous driving systems.