Search

This article investigates the issue of fixed-time path-following control for autonomous ground vehicles (AGVs) with completely unknown parameters subject to preset error constraints. Before converging the path-following errors, a fixed-time prescribed performance control (PPC) protocol is developed to guarantee the selected nonsingular sliding manifold evolving within the preset boundaries. Then, the fixed-time PPC target for the preview error of the AGV can be achieved by satisfying the parameter inequality. Furthermore, the designed adaptive laws can ensure the control accuracy without knowledge of vehicle parameters. Finally, the hardware-in-the-loop tests of multiple maneuvering conditions demonstrate that the proposed fixed-time PPC protocol can provide superior comprehensive path-following performance for the AGV over the conventional fixed-time controller, and the preview error can never violate the prescribed performance constraints.

In this article, a generalized fuzzy subset (GFS) method is proposed to assess the time-varying multistate reliability of the perturbation failure coupling measurement system. First, a perturbation-failure coupling mechanism is designed to define the propagation chain of perturbations so as to integrate all the possible perturbations as the inputs of the GFS method. Second, to assess the time-varying multistate reliability, a GFS reliability model is constructed based on the composite limit state. Furthermore, the concept of the uncertain subset boundary is presented to conduct the reliability assessment via a group of embedded interval type-2 fuzzy sets. To address the deficiency of the GFS reliability model, a data-driven strategy is designed to establish the implicit relation between the limited expert knowledge and the membership function. Finally, the experimental tests are carried out to examine the superiority of the GFS method, and parametric studies are also conducted to study the reliability of the PFCM system.

This paper investigates the problem of accurate voltage tracking control for direct current-direct current (DC-DC) boost converter under unknown system parameters and load. Firstly, uncertainties caused by the perturbation of the inductor, capacitor, input voltage and load are approximated by neural networks. Meanwhile, a finite-time observer is proposed to obtain the estimates of lumped uncertainty without any true parameters of the system. Finally, to improve the convergence of output voltage, a finite-time control scheme is proposed for the DC-DC boost converter. It is proven that all signals of the closed-loop system are bounded and both the estimate errors and tracking errors can converge to a small neighborhood of the origin in finite time. Numerical simulations and real-time experiments are presented to demonstrate the effectiveness and superiority of the proposed controller.

Electric vehicles have the potential to reduce oil dependency and decrease emissions. Electric vehicles with four-wheel independent drive have become a promising vehicle architecture. The yaw moment control has been widely applied to improve the performance of four-wheel independent drive electric vehicles. This study develops a new ideal vehicle model as a reference, which has a logarithmic functional relationship between wheel cornering stiffness and the controlled electric vehicles. A control method based on H∞theory and Moore–Penrose inverses property is used to produce the optimal yaw moment to match the ideal vehicle model. The simulation results show that the yaw speed of the controlled electric vehicles increases with lower speed, and the yaw speed and slip angle decrease with a higher speed. The steering performance is efficiently improved when the four-wheel independent drive electric vehicles can accurately match the ideal vehicle model.