Fault-Tolerant Consensus of Multiagent Systems With Prescribed Performance
研究了多智能体系统在未知时变执行器故障和扰动下的容错一致性问题，首次同时考虑乘性和加性时变故障，通过自适应律估计故障信息，保证相邻智能体位置误差在用户定义的时变性能界内。
This article studies the fault-tolerant consensus problem with the guaranteed transient performance of multiagent systems (MASs) subject to unknown time-varying actuator faults and disturbances. The general actuator faults, including both multiplicative and additive time-varying faults, are considered in such a problem for the first time. Both single-integrator modeled agents and double-integrator modeled agents are investigated. The transient performance is ensured in the sense that position errors between each pair of neighboring agents are guaranteed within certain user-defined time-varying performance bounds. Adaptive laws are designed to estimate information about faults and disturbances. For MASs with additive faults, the proposed controllers ensure errors asymptotically converge to zero with guaranteed transient performance. For MASs with both multiplicative faults and additive faults, the proposed controllers ensure errors converge to a residual set without asymptotic convergence but still with guaranteed transient performance. Two simulation examples are provided to evaluate the proposed schemes.