Finite-Time Multilane Fusion Control for 2-D Plane Vehicle Platoon With Sensor and Actuator Faults
针对二维平面车辆队列在速度传感器和执行器故障下的多车道融合问题，设计了有限时间容错控制器，并引入改进的预设性能函数处理完全执行器故障，确保队列稳定。
This article focuses on the 2-D plane finite-time multilane fusion control problem with velocity sensor faults and actuator faults. First, considering the velocity sensor faults, radial basis function neural networks (RBFNNs) are introduced to approximate sensor fault functions. A finite-time fault-tolerant position controller is designed by employing a hyperbolic tangent function to address the partial failure of the position actuator. Second, in case of complete actuator failure, the backup actuator is activated without altering the controller structure. The majority of prescribed performance functions (PPFs) currently in use are unsuitable for addressing complete actuator failure. As tracking errors may exceed PPF constraints before the backup actuator is activated upon such faults, this could result in vehicle platoon instability. To address this issue, this article designs a modified PPF (MPPF) that is independent of the initial conditions and can adjust the performance boundary according to the error change at specific times by introducing a shifting function. When a complete actuator failure occurs, the MPPF can sacrifice part of the transient performance to enclose the increased tracking error within the range of the MPPF, thus maintaining the stability of the vehicle platoon. When the actuator is working normally or has a partial failure, it can restore the user-specified performance. Then, by constructing a finite-time angle sliding-mode surface, an angle controller is designed. The designed position and angle of finite-time controllers can ensure that the vehicle platoon achieves multilane fusion within a finite time. Finally, through simulation and comparative results, the effectiveness of the proposed MPPF and finite-time fault-tolerant algorithm is demonstrated.