Neural Network-Based Distributed Finite-Time Tracking Control of Uncertain Multi-Agent Systems With Full State Constraints

Abstract

This paper addresses the distributed tracking control problem of pure-feedback multi-agent systems with full state constraints under a directed graph in finite time. By introducing the nonlinear mapping technique, the system with full state constraints is converted into the form without state constraints. Furthermore, by combining fractional dynamic surface and radial basis function neural networks, a novel finite-time adaptive tracking controller is conducted recursively. In light of Lyapunov stability theory, it is proven that all signals of multi-agent systems are semi-globally uniformly ultimately bounded in finite time and the full states satisfy the constraints. Lastly, numerical simulations are supplied to demonstrate the effectiveness of the proposed control strategy. INDEX TERMS Multi-agent systems, finite-time tracking control, full state constraints, pure-feedback form.

Adaptive Practical Fast Finite-Time Consensus Protocols for High-Order Nonlinear Multi-Agent Systems With Full State Constraints

“…Remark 3: Assumption 2 is reasonable and similar to the one proposed for the strict-feedback systems with output constraints in [52], and when Ω Lj = Ω Hj = 0 it reduces to the time-invariant constraints considered in [47].…”

“…where Ω Lj (t) and Ω Hj (t) are known positive functions. Remark 1: The non-strict-feedback system (1) that we considered which is much practical significance represents a wide range of systems, including systems with strictfeedback form [46] and pure-feedback form [47] as special cases. Videlicet, all the system functions g i,j (x i ) and f i,j (x i ) involve the whole state variable, i.e., x i = (x i,1 , .…”

“…Yet, it is worth noting that these researches focus on output constrained systems, other states are unconstrained, while in realistic practical problems, all states of many systems are bounded. More recently, several results occurred about the consensus problem for nonlinear MASs with state constraints [45]- [47]. Especially, in [45] compare with other research about dynamic surface control (DSC) technique, Zhao et al introduce a different NM transformation with the aim of the nonlinear state-dependent function for backstepping design based on DSC to solve the consensus problem with state constraints.…”

“…Based on a high-gain observer, adaptive back-stepping control approach and dynamic surface control technique, and combining BLF method, a consensus tracking control issue of nonlinear strict-feedback MASs with state constraints was investigated in [46]. And Ji et al introduced the fractional dynamic surface method and NM method to investigate the FTC problem of a class of pure-feedback MASs with time-invariant state constraints under directed graph [47]. However, to the authors' best knowledge, there is no result concerned with FTC design for non-strict-feedback MASs with time-varying asymmetric state constraints, which motives the present research.…”

Reinforcement learning‐based adaptive consensus control for a class of non‐affine nonlinear multiagent systems with semi‐Markov switching topographies

“…In many studies of multiagent control methods, most of the controlled systems are assumed to be affine systems. However, in engineering practice, many systems have non‐affine properties, such as mechanical systems, chemical systems and flight control systems, and their control method differs from that of Affine system, which has been published in References 34‐36. Therefore, in this paper, the influencing factors mentioned above are considered in the design process of the controller.…”