Risk-Aware High-level Decisions for Automated Driving at Occluded Intersections with Reinforcement Learning
This addresses safer automated driving at intersections, but it is incremental as it builds on existing DQN methods with risk-aware modifications.
The paper tackled the problem of learning high-level driving decisions at unsignalized occluded intersections using reinforcement learning, proposing a risk-aware DQN approach that outperformed a collision-based DQN and a rule-based policy by providing safer actions and being less overcautious.
Reinforcement learning is nowadays a popular framework for solving different decision making problems in automated driving. However, there are still some remaining crucial challenges that need to be addressed for providing more reliable policies. In this paper, we propose a generic risk-aware DQN approach in order to learn high level actions for driving through unsignalized occluded intersections. The proposed state representation provides lane based information which allows to be used for multi-lane scenarios. Moreover, we propose a risk based reward function which punishes risky situations instead of only collision failures. Such rewarding approach helps to incorporate risk prediction into our deep Q network and learn more reliable policies which are safer in challenging situations. The efficiency of the proposed approach is compared with a DQN learned with conventional collision based rewarding scheme and also with a rule-based intersection navigation policy. Evaluation results show that the proposed approach outperforms both of these methods. It provides safer actions than collision-aware DQN approach and is less overcautious than the rule-based policy.