带有智能领导者的网络系统分群投影一致性

针对多智能体系统在分群结构下投影一致性问题，引入了一种新的领导者称之为智能领导者。与传统的领导者相比，智能领导者可以获取和利用相邻跟随者位置信息，这些信息是否会用于调整智能领导者的控制算法由事件触发函数决定，从而很大程度上提高了系统的鲁棒性和容错性。设计了一个新颖的分布式控制协议；利用矩阵理论和Lyapunov稳定性理论，给出了系统在分群结构下实现投影一致性的充分条件；基于拉普拉斯变换性质，分析了当一些跟随者发生执行器故障时，系统能够保证跟踪误差有界；最后数值仿真验证了结果的有效性。

Projective Group Consensus of Network Systems with Smart Leader

Aiming at projective consensus problem of multi-agent systems under the cluster structure, a new kind of leader is introduced called the smart leader. Compared with traditional leader, where the leader can obtain and utilize its neighboring followers’position information, and whether such information will be used in adjusting the control algorithm for the smart leader is decided by an event-trigger function, thus the robustness and fault tolerance of the system are improved to a great extent. Novel distributed control protocols are designed, then by using matrix theory and Lyapunov stability theory, sufficient conditions are given to achieve the projective consensus of the systems under cluster structure. In addition, based on the Laplace transform property, the system can ensure the tracking error to be bounded when actuator faults occur on some of followers. Finally, numerical simulation verifies the validity of the results.