Motion and formation control of a class of multi-agent systems with adaptive internal model based output regulation method

In this thesis the problem of moving in a predefined formation along a reference trajectory in a class of multi-agent systems in which the agents have general nonlinear dynamics is considered. The problem is formulated as an output regulation problem. Under the assumption that the agent dynamics and the systems that generate the reference trajectories have uncertain parameters, an adaptive internal model based control method is used to solve the problem. At first the problem of moving in a predefined formation along a periodic trajectory is addressed. Then performing various formation maneuvers such as expansion/contraction, rotation, and reconfiguration (change of geometric shape) along a periodic trajectory is studied. As the next step, tracking more complex trajectories is provided by using switching in the systems which generate the periodic reference trajectories. Finally, the problem of distributed agreement is addressed with using different neighborhood topologies. Ring, fully connected, nearest neighbors and random neighborhood topologies are used to determine the agent neighborhoods. The performance of the method for the solution of the problems is verified with numerical simulations.