Local trajectory planning and control for autonomous vehicles using the adaptive potential field

Abstract

In this paper, a new potential field approach is proposed for trajectory planning and control in autonomous vehicles. The potential field of the surrounding environment is generated including vehicles, road boundaries and lane centers. Based on the predicted positions of the vehicles, the location of the ego vehicle and the surrounding potentials are synchronized. In addition, the potential fields of the surrounding vehicles are adaptively modified in shape depending on the relative velocity of the surrounding vehicles. The longitudinal distance required for the lateral avoidance is mathematically calculated and reflected in the potential field. Based on the proposed potential field, the trajectory of the autonomous vehicle is selected as the suboptimal path and the MPC (Model Predictive Control) method is applied for tracking control and the lateral stability of the vehicle. The performance of the proposed algorithm is verified in simulations under various conditions.