Iterative Learning Control for Nonlinear Uncertain Parameterized Multi-Agent Systems With Non-Identical Partially Unknown Control Directions

This paper investigates the learning consensus control and formation control of nonlinear uncertain parameterized multi-agent systems with non-identical partially unknown control directions, which generalizes the research results of linear uncertain parameterized multi-agent systems. Based on neural networks and Fourier series expansion, a new neural network approximator is constructed to approximate the nonlinear uncertain parameterized dynamics. Combined with the iterative learning control and adaptive control method, a novel adaptive iterative learning control law is designed, in which the parameter adaptive iterative learning estimation is used to eliminate the influence of approximation errors, unpredictable leader dynamics and unknown control directions. Then, a new parameterized composite energy function is constructed to demonstrate the stability of the entire consistent systems and formation systems. On this basis, the advantages of protocols in undirected and directed topologies were discussed. Finally, the simulation results verify the effectiveness of the proposed control algorithm.