Research on trajectory tracking control of driverless cars based on game theory

Abstract

The purpose of this game theory-based trajectory tracking control study of driverless cars is to resolve the conflicting problems of trajectory precision in tracking and drive stability for driverless cars in lane change situations. The general plan for control is made. The lateral control is based on the theory of evolutionary games, and the linear quadratic regulator (LQR) is a method for linear quadratic control with predictive feedforward. When it comes to dynamic systems that vary over time, trajectory tracking precision and drive stability are both sides of the same coin. The payoff matrix is first constructed to determine the utility function, followed by the dynamical replication system to evolve the weights of both sides, and finally the optimal an equilibrium strategy for weight allocation between the two sides of the game to achieve the optimal objective function is determined. The longitudinal dual PID controller has been designed based on proportional, differential, and integral theory. The results reveal that the developed controller outperforms the LQR controller in terms of tracking outcomes, as well as path tracking precision and drive stability.