Robust Decision Making for Autonomous Vehicles at Highway On-Ramps: A Constrained Adversarial Reinforcement Learning Approach

Reinforcement learning has demonstrated its potential in a series of challenging domains. However, many real-world decision making tasks involve unpredictable environmental changes or unavoidable perception errors that are often enough to mislead an agent into making suboptimal decisions and even cause catastrophic failures. In light of these potential risks, reinforcement learning with application in safety-critical autonomous driving domain remains tricky without ensuring robustness against environmental uncertainties (e.g., road adhesion changes or measurement noises). Therefore, this paper proposes a novel constrained adversarial reinforcement learning approach for robust decision making of autonomous vehicles at highway on-ramps. Environmental disturbance is modelled as an adversarial agent that can learn an optimal adversarial policy to thwart the autonomous driving agent. Meanwhile, observation perturbation is approximated to maximize the variation of the perturbed policy through a white-box adversarial attack technique. Furthermore, a constrained adversarial actor-critic algorithm is presented to optimize an on-ramp merging policy while keeping the variations of the attacked driving policy and action-value function within bounds. Finally, the proposed robust highway on-ramp merging decision making method of autonomous vehicles is evaluated in three stochastic mixed traffic flows with different densities, and its effectiveness is demonstrated in comparison with the competitive baselines.

Automated summary

This paper proposes a novel constrained adversarial reinforcement learning approach for robust decision making of autonomous vehicles at highway on-ramps. Environmental disturbance is modelled as an adversarial agent and observation perturbation is approximated to maximize the variation of the perturbed policy. A constrained adversarial actor-critic algorithm is presented to optimize an on-ramp merging policy while keeping the variations of the attacked driving policy and action-value function within bounds. The effectiveness of the proposed method is demonstrated in three stochastic mixed traffic flows with different densities compared to competitive baselines.