Virtual decomposition with time delay control for underactuated robot manipulator

Abstract

The importance of controlling robot manipulators is undeniable. However, faults in these systems can significantly impact the workspace environment and personal safety. To address these challenges, a new adaptive approach has been proposed that easily adapts to a faulty actuator while precisely tracking its desired position. The virtual decomposition control (VDC) method decomposes the robot into subsystems, each with its sub-controller, while ensuring the overall system remains stable. Meanwhile, time delay estimation (TDE) is used to estimate unknown and uncertain parameters. A co-simulation was conducted to test the TD-VDC method on a 6 DoFs robot, which becomes underactuated during its running. The results of the root main square error of the proposed controller were lower of 6% than those of sliding mode control based on partial feedback linearization control (SMC-PFLC), which proves the proposal's effectiveness and efficiency.