Replanning-Based Nonlinear Model Predictive Tracking Control for Intelligent Vehicle on Unexpected Slippery Road

Conventional path tracking controllers strictly follow preplanned trajectories, which can become impractical or even infeasible under unexpected low-adhesion conditions. This often results in conflicts between precise tracking requirements and the vehicle’s real-time motion capabilities. In particular, designing tracking controllers for slippery road conditions involves addressing multiple constraints and complex control objectives. To overcome these challenges, this article introduces a replanning-based nonlinear model predictive path tracking control strategy that dynamically adjusts the tracking path to account for various constraints and objectives. A simplified quasi-linear tire force slip model is utilized to compute the front wheel steering angle directly, enhancing computational efficiency. An explicit reference path replanning module is developed to dynamically modify target trajectories, integrating multiple constraints and path tracking objectives. Moreover, a weight-scaling nonlinear model predictive controller is proposed to operate alongside the replanning module, enabling effective path tracking under varying conditions and conflicting objectives. Extensive hardware-in-the-loop (HIL) experimental results validate the effectiveness and superiority of the proposed control strategy.