Fuzzy Logic-Enhanced Neuroadaptive Fault-Tolerant Control for Vehicular Platoons With Stochastic Disturbances and Asymmetric Spacing Constraints

This article introduces a novel fuzzy logic-enhanced neuroadaptive sliding mode control (FLENNSMC) framework, developed for vehicular platoon systems subject to a confluence of challenges. Leveraging the synergistic integration of fuzzy logic's interpretive strengths and neural networks' adaptive learning capabilities, FLENNSMC effectively addresses nonlinear dynamics, stochastic disturbances, actuator faults, and stringent asymmetric spacing constraints. We propose a Takagi-Sugeno (T-S) fuzzy model to structure the learning process and a fuzzy logic-enhanced RBFNN (FLERBFNN) for robust approximation of unknown functions, including unmodeled dynamics and fault signals. The controller design incorporates a fault-tolerant control mechanism for enhanced robustness, an asymmetric barrier Lyapunov function (BLF) to strictly enforce spacing constraints, and a Nussbaum function to compensate for actuator faults with unknown directions. The fuzzy logic-enhanced structure allows for localized and efficient learning, which reduces computational burden and improves adaptation speed. Through a rigorous stochastic Lyapunov-Krasovskii stability analysis, we derive sufficient LMI-based conditions for the uniform ultimate boundedness (UUB) of tracking errors in the mean square sense and guarantee a mixed H-infinity/passivity performance. Extensive simulations on a 2-D multilane vehicular platoon demonstrate the superior performance of the proposed FLENNSFC compared to conventional neuroadaptive control approaches, particularly highlighting the benefits of fuzzy logic in structuring the learning process and handling complex uncertainties. Simulation code is available at https://github.com/zhanganguo/FLENNSMC-Platoon-Control-Simulation.