Switching Dynamic Event-Triggered Prescribed Performance Control for Underactuated ASVs
Ying Zhao, Wenlong Lu, Yuqing Chen
Abstract
In this paper, a switching dynamic event-triggered prescribed performance formation control (SDPPC) algorithm is proposed for underactuated autonomous surface vehicles (ASVs). The control objective is to achieve high-precision formation control tasks for the ASVs while ensuring transient performance, connectivity, collision avoidance, and communication resource conservation. Collision avoidance and connectivity maintenance are achieved by defining novel predefined performance functions. The radial basis function neural network (RBF-NN) is utilized to approximate the uncertainty of the unknown model dynamics in the form of a minimum learning parameter (MLP). The ‘algebraic loop’ issue frequently encountered in the back-stepping method is resolved by the inherent properties of the radial basis function. A novel switching dynamic event-triggered mechanism (SDETM) is developed, which can save the communication resources in case of oscillations and satisfy the high-precision formation control task. By utilizing Barbalat's Lemma and the Lyapunov function method, it has been proven that the formation tracking errors can converge to zero as time tends towards infinity. Finally, the feasibility and effectiveness of the SDPPC algorithm are demonstrated through numerical simulations.