In this paper, neuroadaptive fault-tolerant control of nonlinear vehicular platoon systems subject to unknown direction actuator faults, unmodeled dynamics, and external disturbances is investigated. A vehicle model considering the influence of velocity direction deflection angle on vehicle position is proposed on a two-dimensional (2-D) plane to realize multi-lane vehicle fusion. Then, in order to avoid collisions and maintain communication connection, the method of asymmetric barrier Lyapunov function (BLF) is adopted to eliminate the unfavorable assumption on spacing constraints in using symmetric BLF in the existing result. Nussbaum function is adopted to attenuate the negative effects caused by unknown direction actuator faults. Furthermore, by combining sliding-mode control (SMC) techniques with radial basis function neural network (RBFNN), a novel neuroadaptive fault-tolerant control scheme with minimal learning parameters is designed to not only guarantee the finite-time stability of the whole vehicular platoon but also tolerate the unknown direction actuator faults. Finally, simulation results show the effectiveness and advantages of the proposed scheme.

本文研究了非线性车辆排系统在未知方向执行器故障，未建模动力学和外部干扰作用下的神经自适应容错控制。在二维（2-D）平面上提出了考虑速度方向偏转角对车辆位置影响的车辆模型，以实现多车道车辆融合。然后，为了避免冲突并保持通信连接，在现有结果中采用非对称势垒李雅普诺夫函数（BLF）的方法消除了关于使用对称BLF的间距约束的不利假设。采用Nussbaum函数可减轻未知方向致动器故障引起的负面影响。此外，通过将滑模控制（SMC）技术与径向基函数神经网络（RBFNN）结合起来，设计了一种具有最小学习参数的新型神经自适应容错控制方案，不仅可以保证整个车辆排的有限时间稳定性，而且可以容忍未知方向执行器故障。最后，仿真结果表明了该方案的有效性和优势。