Risk-Constrained Interactive Safety under Behavior Uncertainty for Autonomous Driving
This work provides an interpretable and tunable safety objective for autonomous driving systems operating in dense traffic, addressing the trade-off between safety and efficiency for developers and users.
This paper addresses the challenge of balancing safety and efficiency in autonomous driving within dense traffic by adopting a probabilistic risk objective for interactive planning. They formalize a Risk-Constrained Robust Stochastic Bayesian Game and solve it using a variant of Multi-Agent Monte Carlo Tree Search, demonstrating superior performance over baselines in simulation while maintaining a specified and interpretable risk level.
Balancing safety and efficiency when planning in dense traffic is challenging. Interactive behavior planners incorporate prediction uncertainty and interactivity inherent to these traffic situations. Yet, their use of single-objective optimality impedes interpretability of the resulting safety goal. Safety envelopes which restrict the allowed planning region yield interpretable safety under the presence of behavior uncertainty, yet, they sacrifice efficiency in dense traffic due to conservative driving. Studies show that humans balance safety and efficiency in dense traffic by accepting a probabilistic risk of violating the safety envelope. In this work, we adopt this safety objective for interactive planning. Specifically, we formalize this safety objective, present the Risk-Constrained Robust Stochastic Bayesian Game modeling interactive decisions satisfying a maximum risk of violating a safety envelope under uncertainty of other traffic participants' behavior and solve it using our variant of Multi-Agent Monte Carlo Tree Search. We demonstrate in simulation that our approach outperforms baselines approaches, and by reaching the specified violation risk level over driven simulation time, provides an interpretable and tunable safety objective for interactive planning.