Hybrid variable admittance force tracking and fixed-time position control for robot–environment interaction

Abstract

Purpose This paper aims to deal with the force and position tracking problem when a robot performs a task in interaction with an unknown environment and presents a hybrid control strategy based on variable admittance control and fixed-time control. Design/methodology/approach A hybrid control strategy based on variable admittance control and fixed-time control is presented. Firstly, a variable stiffness admittance model control based on proportional integral and differential (PID) is adopted to maintain the expected force value during the task execution. Secondly, a fixed-time controller based on radial basis function neural network (RBFNN) is introduced to handle the model uncertainties and ensure the fast position tracking convergence of the robot system, while the singularity problem is also avoided by designing the virtual control variable with piecewise function. Findings Simulation studies conducted on the robot manipulator with two degrees of freedom have verified the superior performance of the proposed control strategy comparing with other methods. Originality/value A hybrid control scheme for robot–environment interaction is presented, in which the variable stiffness admittance method is adopted to adjust the interaction force to the desired value, and the RBFNN-based fixed-time position controller without singularity problem is designed to ensure the fast convergence of the robot system with model uncertainty.