Distributed Cooperative Path-Planning for Autonomous Vehicles Integrating Human Driver Trajectories

This paper considers the problem of cooperative optimal trajectory generation for autonomous driving involving a human driver vehicle (HDV) into the traffic scenario. A distributed model predictive control (MPC) approach to solve the cooperative path-planning problem is proposed. The novelty of this paper are in to enhance cooperative path-planning algorithms, that allows for cooperation between autonomous and human-driven vehicles. We integrate the trajectory prediction of the human driver model (HDM) into the framework, such that the human vehicle influence the behaviour of the other agents. Mixed-integer quadratic programming (MIQP) it is appealing to solve the cooperative planning. The proposed method is evaluated by computational simulations, which show the effectiveness of our approach for a cooperative overtaking scenario comprising three vehicles in a three-lane, one way road.