Semi-Parametric Control Architecture for Autonomous Underwater Vehicles Subject to Time Delays

This paper presents a data-driven model-based control system for autonomous underwater vehicles (or AUVs) subject to input delays. work is motivated by the time delays that can arise in robotics due communication restrictions and sensor malfunctions. Such highly degrade performance of classical structures resulting unpredictable behaviours. The proposed architecture addresses such limitations. approach incorporates linear dynamic representation obtained using Koopman operator an observer/state prediction formulation. designed based on discrepancies between estimation system’s behaviour actual AUV chain predictors. capabilities are shown through experiments performed with 4 degrees-of-freedom vehicle. results demonstrate stable behaviours without steady-state errors, even presence long delay.