Autonomous vehicles regardless of the drivetrain configuration are highly sensitive to disturbances, uncertain dynamic parameters, and modeling errors. Neglecting these factors during trajectory-tracking or lane-keeping can cause the autonomous vehicle (AV) to deviate from its reference path, compromising safety and performance. In this work, a fixed-time prescribed performance backstepping controller integrated with a super-twisting-like algorithm is proposed to ensure fixed-time convergence of trajectory-tracking errors and robust stability under bounded uncertainty factors and external disturbances. A fixed-time prescribed performance approach is utilized to constrain the evolution of lateral and angular tracking errors, thereby limiting the risk of divergence and ensuring control stability. This framework is demonstrated by the Lyapunov-based stability analysis to demonstrate fixed-time stability in an arbitrarily small neighborhood around the origin. The framework is also validated through simulation on full-scale vehicle model. Moreover, virtual hardware-in-the-loop and real-time experiments are conducted on a reduced-scale QCar platform under uncertain parameters and external disturbances.
Bancel et al. (2026) studied this question.