Fully-Distributed Leader-Follower Flocking Protocol for Time-Depend Lagrange systems with External Disturbance

This paper designs a protocol for leader-follower flocking problem, for multi-agent time-depend Lagrange systems (MATDLS) with external disturbances problem. The mechanical systems with varying mass can be modeled with time-depend Lagrange systems. The variable mass causes to uncertainty in the model. We suppose that the virtual leader has a bounded acceleration, and only its neighbor agents can receive its information. The study of leader-follower flocking with dynamic leader/virtual leader (LFFDL) for MATDLS under external disturbance problem is very challengeable. An adaptive estimator is allocated to each agent for estimation of the leader's velocity, and an adaptive control is proposed to solve LFFDL problem. The benefits of our protocol are both fully-distributed and continuous. Under the proposed control protocol, each follower's velocity asymptotically tends to that of the virtual leader, the group’s network stays connected, and no collision happens between agents. In the end, we perform some simulations to show the theoretical results validation.