Fully Distributed Optimization Algorithm for Nonlinear Multiagent Systems Under Stochastic Event-Triggered Mechanism

This article focuses on the distributed optimization problem of heterogeneous nonlinear multiagent systems with coupling constraints. By decoupling the design, the problem is transformed into two components: the optimal signal estimator design with coupling constraints and the tracking controller design with nonlinearity. To achieve this, a novel optimal signal estimator is proposed, which effectively manages coupling constraints without requiring global information. The estimator incorporates adaptive gains, a crucial element for achieving a fully distribution algorithm. In addition, a correction term is introduced in the adaptive gain update law to filter out high-frequency signals and enhance the algorithm’s robustness. To further reduce communication frequency, a stochastic event-triggered mechanism is employed, allowing agents to trigger with a certain probability. Using the optimal signal and its higher order derivative trajectories generated by the estimator as reference signals, a tracking controller is constructed through backstepping adaptive technique to ensure asymptotic tracking and effectively solve the distributed optimization problem. Finally, the effectiveness of the optimization algorithm is validated through simulation.