This paper addresses the finite-time leader-following output-feedback consensus problem for a class of high-order nonlinear multi-agent systems. The agents dynamics are supposed to be in lower-triangular form and satisfy Lipschitz conditions with time-varying gains. Based on the dynamic output approach and dynamic gain control method, a new distributed output feedback consensus protocol is proposed with two online tuned gains such that the finite-time leader-following consensus is achieved. Different from some exiting observer-based output feedback approaches that the observers embedded in agents have to share full observation information with their neighbors, the proposed protocol only requires the output information of neighboring agents to be transmitted, and thus the network communication burden is reduced. In addition, compared to the traditional backstepping method that commonly used for finite time control of nonlinear system, the proposed controller can avoid the repeated derivation problem in backstepping control and thereby is simple and convenient for use. Finally, a numerical example is given to demonstrate the effectiveness of the proposed results.

本文针对一类高阶非线性多智能体系统，解决了有限时间领导者跟随输出反馈共识问题。代理的动力学应该是低三角形的，并且可以满足Lipschitz条件，并且具有随时间变化的收益。基于动态输出方法和动态增益控制方法，提出了一种新的分布式输出反馈共识协议，该协议具有两个在线调整的增益，从而实现了有限时间的领导者遵循共识。与某些退出的基于观察者的输出反馈方法不同，嵌入在代理中的观察者必须与邻居共享完整的观察信息，所提出的协议仅要求传输相邻代理的输出信息，从而减轻了网络通信负担。此外，与通常用于非线性系统有限时间控制的传统反步方法相比，该控制器可以避免反步控制中的重复推导问题，使用简单方便。最后，通过数值例子说明了所提出结果的有效性。