arXiv reaDer
Transient Fault Tolerant Semantic Segmentation for Autonomous Driving
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.
updated: Fri Aug 30 2024 00:27:46 GMT+0000 (UTC)
published: Fri Aug 30 2024 00:27:46 GMT+0000 (UTC)
参考文献 (このサイトで利用可能なもの) / References (only if available on this site)
被参照文献 (このサイトで利用可能なものを新しい順に) / Citations (only if available on this site, in order of most recent)
Amazon.co.jpアソシエイト