Self-Improving Safety Performance of Reinforcement Learning Based Driving with Black-Box Verification Algorithms

Creative Commons License

Dagdanov R., Durmuş H., Üre N. K.

2023 IEEE International Conference on Robotics and Automation, ICRA 2023, London, England, 29 May - 02 June 2023, vol.2023-May, pp.5631-5637 identifier identifier

  • Publication Type: Conference Paper / Full Text
  • Volume: 2023-May
  • Doi Number: 10.1109/icra48891.2023.10160883
  • City: London
  • Country: England
  • Page Numbers: pp.5631-5637
  • Keywords: Autonomous Driving, Black-Box Verification, Deep Reinforcement Learning
  • Istanbul Technical University Affiliated: Yes


In this work, we propose a self-improving artificial intelligence system to enhance the safety performance of reinforcement learning (RL)-based autonomous driving (AD) agents using black-box verification methods. RL algorithms have become popular in AD applications in recent years. However, the performance of existing RL algorithms heavily depends on the diversity of training scenarios. A lack of safety-critical scenarios during the training phase could result in poor generalization performance in real-world driving applications. We propose a novel framework in which the weaknesses of the training set are explored through black-box verification methods. After discovering AD failure scenarios, the RL agent's training is re-initiated via transfer learning to improve the performance of previously unsafe scenarios. Simulation results demonstrate that our approach efficiently discovers safety failures of action decisions in RL-based adaptive cruise control (ACC) applications and significantly reduces the number of vehicle collisions through iterative applications of our method. The source code is publicly available at