Convergence of multi-agent systems controlled by iterative learning strategies with continuous data losses

To address the consistency tracking issue of multi-agent systems operating repetitive tasks with continuous data losses, this paper proposes an iterative learning control (ILC) strategy with data compensation. The convergence of multi-agent ILC systems is demonstrated by analyzing changes in the element values of the transition matrix of input errors. It is worth noting that the analysis also reveals two useful conclusions. One is that the convergence speed decreases as the probability of continuous data loss increases. The other is that the selection of the controller’s learning gain depends on the maximum value of the number of continuous data losses. Simulation results verify the effectiveness of the proposed ILC strategy and the theoretical derivations regarding convergence speed and gain selection.