Distributed Linear Parameter Varying Model Predictive Controller with Event-Triggered Mechanism for Nonholonomic Mobile Robot

Abstract

This paper deals with the design of distributed linear parameter varying model predictive controller (LPVMPC) to solve consensus problem on nonholomonic model with event-triggered mechanism. The nonlinear dynamic of nonholomonic model is simplified into linear time variant model by discretizing with Euler method. A quadratic cost function is determined by including terminal state and varying weight matrices, in order to reduce offset due to modeling simplification. The control signal for very agents are calculated by solving quadratic programming problem. A local optimal state controller is integrated with LPV-MPC during eventtriggered mechanism. In order to reduce computational load, the predictive controller only performs optimization only when the trigger conditions are met. The proposed controller is also verified in case of numerical simulation test, and shown its capability to provide good response of nonholomonic mobile robot’s consensus protocol.