# Event-triggered iterative learning control for output constrained multi-agent systems

## TL;DR

This paper introduces a control strategy for multi-agent systems that reduces communication by using event-triggered learning, ensuring accurate tracking of desired outputs.

## Contribution

The novelty lies in the event-triggered iterative learning control algorithm that enables consensus tracking without real-time communication.

## Key findings

- The proposed algorithm ensures convergence using a Lyapunov function.
- Simulation results confirm the effectiveness of the control protocol.
- The method allows consistent tracking of desired trajectories with output constraints.

## Abstract

An event-triggered iterative learning consensus tracking control strategy is proposed for output constrained nonlinear discrete-time multi-agent systems. Firstly, the estimated Pseudo partial derivative(PPD) algorithm is determined based on the input and output data of the system, and the output observer is designed based on the estimated PPD. Secondly, the deadband controller is designed based on the output estimation error of the observer, and the event trigger condition is determined by comparing the size of the output estimation error and the deadband controller function value, and the agents communicate when the trigger condition is satisfied, and do not communicate when it is not satisfied. Then, the event-triggered iterative learning control algorithm is constructed using the estimated PPD, the trigger condition and the measurement error, and the convergence of the algorithm is proved by using the Lyapunov function, and the proposed algorithm can make the output constrained multi-agent system consistently and completely tracking on the desired trajectory without the need of real-time communication conditions. Finally, the simulation results further validate the effectiveness of the control protocol.