This paper addresses a formation tracking problem for nonlinear multi-agent systems with time-varying actuator faults, in which only a subset of agents has access to the leader's information over the directed leader-follower network with a spanning tree. Both the amplitudes and signs of control coefficients induced by actuator faults are unknown and time-varying. The aforementioned setting improves the practical relevance of the problem to be investigated, and meanwhile, it poses technical challenges to distributed controller design and asymptotic stability analysis. By introducing a distributed estimation and control framework, a novel distributed control law based on a Nussbaum gain technique is developed to achieve robust fault-tolerant formation tracking for heterogeneous nonlinear multi-agent systems with time-varying actuator faults. It can be proved that the asymptotic convergence is guaranteed. In addition, the proposed approach is applied to task-space cooperative tracking of networked manipulators irrespective of the uncertain kinematics, dynamics, and actuator faults. Numerical simulation results are presented to verify the effectiveness of the proposed designs.

中文翻译：

---

具有时变执行器故障的异构多智能体系统的容错编队跟踪及其在机械手任务空间协同跟踪中的应用

本文解决了具有时变执行器故障的非线性多智能体系统的编队跟踪问题，其中只有一部分智能体可以通过具有生成树的有向领导者-跟随者网络访问领导者的信息。由执行器故障引起的控制系数的幅度和符号都是未知的并且随时间变化。上述设置提高了待研究问题的实际相关性，同时，它对分布式控制器设计和渐近稳定性分析提出了技术挑战。通过引入分布式估计和控制框架，开发了一种基于 Nussbaum 增益技术的新型分布式控制律，以实现具有时变执行器故障的异构非线性多智能体系统的鲁棒容错地层跟踪。可以证明渐近收敛是有保证的。此外，无论运动学、动力学和执行器故障如何，所提出的方法都适用于网络机械手的任务空间协同跟踪。数值模拟结果被提出来验证所提出的设计的有效性。