We propose a methodology for designing dependable Artificial Neural Networks (ANN) by extending the concepts of understandability, correctness, and validity that are crucial ingredients in existing certification standards. We apply the concept in a concrete case study in designing a high-way ANN-based motion predictor to guarantee safety properties such as impossibility for the ego vehicle to suggest moving to the right lane if there exists another vehicle on its right.
@article{arxiv.1709.00911,
title = {Neural Networks for Safety-Critical Applications - Challenges, Experiments and Perspectives},
author = {Chih-Hong Cheng and Frederik Diehl and Yassine Hamza and Gereon Hinz and Georg Nührenberg and Markus Rickert and Harald Ruess and Michael Troung-Le},
journal= {arXiv preprint arXiv:1709.00911},
year = {2017}
}
Comments
Summary for activities conducted in the fortiss Eigenforschungsprojekt "TdpSW - Towards dependable and predictable SW for ML-based autonomous systems". All ANN-based motion predictors being formally analyzed are available in the source file