Multi-Agent Systems are susceptible to external disturbances, sensor failures or collapse of communication channel/media. Such failures disconnect the agent network and thereby hamper the consensus of the system. Quick recovery of consensus is vital to continue the normal operation of an agent-based system. However, only limited works in the past have investigated the problem of recovering the consensus of an agent-based system in the event of a failure. This work proposes a novel algorithmic approach to recover the lost consensus, when an agent-based system is subject to the failure of an agent. The main focus of the algorithm is to reconnect the multi-agent network in a way so as to increase the connectivity of the network, post recovery. The proposed algorithm may be applied to both linear and non-linear continuous-time consensus protocols. To verify the efficiency of the proposed algorithm, it has been applied and tested on two multi-agent networks. The results, thus obtained, have been compared with other state-of-the-art recovery algorithms. Finally, it has been established that the proposed algorithm achieves better connectivity and therefore, faster consensus when compared to the other state-of-the-art.

中文翻译：

---

代理失效多代理系统的最优共识恢复

多代理系统容易受到外部干扰、传感器故障或通信通道/媒体崩溃的影响。此类故障会断开代理网络，从而阻碍系统的共识。快速恢复共识对于继续基于代理的系统的正常运行至关重要。然而，过去只有有限的工作研究了在发生故障时恢复基于代理的系统的共识问题。这项工作提出了一种新的算法方法来恢复丢失的共识，当基于代理的系统受到代理失败的影响时。该算法的主要重点是以某种方式重新连接多代理网络，以增加网络的连通性，后恢复。所提出的算法可以应用于线性和非线性连续时间一致性协议。为了验证所提出算法的效率，它已经在两个多智能体网络上进行了应用和测试。将由此获得的结果与其他最先进的恢复算法进行了比较。最后，已经确定，与其他最先进的算法相比，所提出的算法实现了更好的连通性，因此实现了更快的共识。