Robust cooperative control reconfiguration/recovery in multi-agent systems

An H∞ cooperative fault recovery control of multi-agent systems

“…Consider the representation of an agent subject to presence of faults be specified as in Subsection II-E, and given by the equation (10) or equivalently by the transformed model (15). Figure 1: The schematic of the i-th pinned agent and its nearest neighbor agents j and k, which are not pinned.…”

“…Recently, the control reconfiguration problem of multi-agent systems has been studied in [1], [5], [9], [10], [13], [19], [20], [25]- [27], [29], [32], [33]. In [1], [25], formation flight problem in a network subject to loss of effectiveness (LOE) faults is considered and in [9], [10], [33] the consensus achievement problem in faulty multi-agent systems is studied. In [25], a discrete-event supervisory module is designed to recover the faults that cannot be recovered by the agents using only local recovery solutions.…”

Output Feedback Fault‐Tolerant Control of Heterogeneous Multi‐Agent Systems

“…However, in active FTC approaches, parameters and/or the structure of the nominal controller are reconfigured according to the occurred fault and they can be divided into two general categories, namely with or without a separate fault detection, isolation and identification (FDII) unit in their structure. A class of active FTC approaches requires an FDII unit that gives, in real time, information regarding the place and the severity of the fault and according to this information, the controller is reconfigured [12–19]. Other active FTC approaches are based on an adaptive control structure without requiring a separate FDII unit [20–25].…”

“…In [12,13], a hierarchical cooperative fault accommodation of unmanned aerial vehicles subject to loss of effectiveness actuator faults is considered. In [15,16], H ∞ ‐based reconfigurable control for the consensus problem of linear time‐invariant (LTI) MAS subject to actuator faults is presented. In [14], a consensus problem for LTI MAS is considered against actuator faults using the virtual actuator idea.…”

“…Our proposed approach has the following main distinctions with previous works reported in the literature, namely: (i) the proposed control schemes in [13–17, 20, 22, 24, 31] are only developed for a specific mission for the multi‐agent systems while our proposed approach is mission‐independent and can be easily augmented to the existing nominal controller for each agent, (ii) approaches in [12–19] are based on the existence of the FDII unit, while our approach does not require it, (iii) approaches in [13–17, 20, 22, 24, 32] are based on the state feedback while our approach only requires output measurement, (iv) approaches in [13–17] require an on‐line parameter design and controller reconfiguration, while in our proposed approach, all parameters are designed off‐line, (v) it is assumed in approaches of [13–17, 20, 22, 24, 32–35] that multi‐agent systems are homogeneous, while this assumption is not needed in our proposed approach, and (vi) in this work, the external disturbance and additive actuator faults are considered, while in [13,14,24] external disturbance and in [13–16] additive actuator fault have not been considered. To summarize, Table 1 presents a comparison between the proposed approach and other works in the literature.…”