Non-zero-sum Game Control for Multi-vehicle Driving via Reinforcement Learning

TL;DR

This paper introduces a novel non-zero-sum game framework for multi-vehicle driving that models interactive decision-making as a whole, using reinforcement learning to derive Nash equilibrium strategies for safe and efficient driving.

Contribution

It presents a unified game control framework combining prediction, decision, and control, solved via ADP-based reinforcement learning for multi-vehicle interactions.

Findings

Vehicles learn interactive behaviors like overtaking.

The algorithm achieves safe, efficient, and comfortable driving.

Effective at non-signalized intersections.

Abstract

When a vehicle drives on the road, its behaviors will be affected by surrounding vehicles. Prediction and decision should not be considered as two separate stages because all vehicles make decisions interactively. This paper constructs the multi-vehicle driving scenario as a non-zero-sum game and proposes a novel game control framework, which consider prediction, decision and control as a whole. The mutual influence of interactions between vehicles is considered in this framework because decisions are made by Nash equilibrium strategy. To efficiently obtain the strategy, ADP, a model-based reinforcement learning method, is used to solve coupled Hamilton-Jacobi-Bellman equations. Driving performance is evaluated by tracking, efficiency, safety and comfort indices. Experiments show that our algorithm could drive perfectly by directly controlling acceleration and steering angle. Vehicles could learn interactive behaviors such as overtaking and pass. In summary, we propose a non-zero-sum game framework for modeling multi-vehicle driving, provide an effective way to solve the Nash equilibrium driving strategy, and validate at non-signalized intersections.