Trajectory Tracking Control Using Evolutionary Approaches for Autonomous Driving

Abstract

Capitalizing on the strides in artificial intelligence and the escalating demand for safer and more efficient traffic systems, the investigation unveils a trio of evolutionary algorithms - namely Grey Wolf Optimizer (GWO), Multi-Verse Optimizer (MVO) and Salp Swarm Algorithm (SSA) - in the context of hyperparameter calibration for the Proportional-Integral-Derivative (PID) controller. The PID controller, revered for its classical design and wide industrial adoption, forms the cornerstone of feedback control systems. To exemplify the utility of the proposed algorithms, two distinct trajectory scenarios are employed as target trajectories. Rigorous numerical evaluations, accompanied by graphical analyses, showcase the prowess of these algorithms in steering the trajectory tracking process. The study unfolds novel contributions, rendering an unprecedented application of these optimizers in the PID controller realm while offering a comprehensive scrutiny of their performances.