Nonlinear Model Predictive Path Tracking Control for Autonomous Vehicles Based on Orthogonal Collocation Method

Autonomous vehicles have gained popularity over the past few years. In this paper, we present a nonlinear model predictive control approach for autonomous vehicles to follow the desired path in various driving environments. Instead of utilizing Euler discretization method which is common in the literature related to MPC-based path tracking, our formulated nonlinear optimal control problem is discretized by orthogonal collocation method for higher tracking accuracy. The presented methodology considers the maximization of longitudinal progression along the reference path, target accurate tracking and comfortable ride. The single-track dynamic model combined with a simplified but effective enough tire model is used. Then the proposed controller is tested and compared with the controller based on Euler discretization in two simulation scenarios to validate its effectiveness. The simulation results demonstrate the integration of longitudinal and lateral motion control of our proposed controller and the better performance in tracking accuracy. Real-time performance of the proposed controller is also discussed.