English

Incorporating System-level Safety Requirements in Perception Models via Reinforcement Learning

Robotics 2024-12-05 v1 Machine Learning

Abstract

Perception components in autonomous systems are often developed and optimized independently of downstream decision-making and control components, relying on established performance metrics like accuracy, precision, and recall. Traditional loss functions, such as cross-entropy loss and negative log-likelihood, focus on reducing misclassification errors but fail to consider their impact on system-level safety, overlooking the varying severities of system-level failures caused by these errors. To address this limitation, we propose a novel training paradigm that augments the perception component with an understanding of system-level safety objectives. Central to our approach is the translation of system-level safety requirements, formally specified using the rulebook formalism, into safety scores. These scores are then incorporated into the reward function of a reinforcement learning framework for fine-tuning perception models with system-level safety objectives. Simulation results demonstrate that models trained with this approach outperform baseline perception models in terms of system-level safety.

Keywords

Cite

@article{arxiv.2412.02951,
  title  = {Incorporating System-level Safety Requirements in Perception Models via Reinforcement Learning},
  author = {Weisi Fan and Jesse Lane and Qisai Liu and Soumik Sarkar and Tichakorn Wongpiromsarn},
  journal= {arXiv preprint arXiv:2412.02951},
  year   = {2024}
}