A finite-time disturbance-observer (DO)-based fuzzy sliding mode cooperative backstepping (FDFSM-CB) scheme is proposed, to improve the synchronization performance and reduce its error under disturbance between multiple permanent magnet synchronous traction motors in low-speed urban rail transit (URT). First, each motor is supposed to an agent of multi-agent system (MAS), and the information between multiple motors can be transmitted through a communication topology network, which ensures the consistency of the response of each agent. Then, considering that the load torque of the motor is disturbed during operation, a finite-time DO is proposed to estimate the unknown load disturbance, thus guaranteeing the anti-disturbance ability of the system. Besides, the nonlinear parts of the dynamic model are approximated by fuzzy logic systems (FLSs), and a second-order sliding mode differentiator is designed to avoid the direct derivation of virtual control law and the problem of differential explosion. Finally, the system is proved to be finite-time-stable. The feasibility and effectiveness of the proposed control scheme are verified by the hardware-in-the-loop (HIL) platform.

This paper investigates the event-triggered fixed-time multi-lane fusion control for vehicle platoon systems with distance keeping constraints where the vehicles are spread in multiple lanes. To realize the fusion of vehicles in different lanes, the vehicle platoon systems are firstly constructed with respect to a two-dimensional (2-D) plane. In case of the collision and loss of effective communication, the distance constraints for each vehicle are guaranteed by a barrier function-based control strategy. In contrast to the existing results regarding the command filter techniques, the proposed distance keeping controller can constrain the distance tracking error directly and the error generated by the command filter is coped with by adaptive fuzzy control technique. Moreover, to offset the impacts of the unknown system dynamics and the external disturbances, an unknown input reconstruction method with asymptotic convergence is developed by utilizing the interval observer technique. Finally, two relative threshold triggering mechanisms are utilized in the proposed fixed-time multi-lane fusion controller design so as to reduce the communication burden. The corresponding simulation results also verify the effectiveness of the proposed strategy.

A deep learning-based cooperative fault diagnosis method is proposed in this study. The proposed method aims to improve the reliability of contemporary fault diagnosis methods for the multipermanent magnet synchronous motor (PMSM) drive system of four-wheel-independent-drive electric vehicles (4WID-EVs). Specifically, a novel 1-D signal to 2-D RGB image conversion method is proposed to transform the fault diagnosis task into an image classification problem, and the impressive image classification capability of the convolutional neural network (CNN) is utilized to achieve a highly accurate fault diagnosis function. Following this, CNNs of various architectures are trained using the constructed RGB image dataset, and the fault diagnosis accuracy under different architectures is compared. In addition, the adaptability of the trained network is increased by utilizing two adaptive factors to ensure that the trained CNN can be compatible with all untrained combinations of angular speed and load torque. Furthermore, by combining the designed single fault diagnosis method with directed-graph theory, a comprehensive evaluation criteria-based cooperative diagnosis method is proposed, which further improves the accuracy significantly. The effectiveness of the proposed fault diagnosis method is demonstrated via adequate comparison.