Variable Admittance Control With Virtual Stiffness Guidance for Human-Robot Collaboration

Abstract

Human-robot collaboration (HRC) is a promising solution for expanding the use of robotic systems in unstructured environments and complex processes in various industries. In this paper, a novel variable admittance control (VAC) with virtual stiffness guidance (VSG) is proposed to improve the performance of HRC. The proposed VAC prevents unnecessary changes of the damping parameter by classifying the human intentions in the low-velocity region, which results in smooth movement. Additionally, the VAC with VSG makes the robot actively assist an operator using a virtual spring. Under the proposed VSG scheme, the equilibrium position of the virtual spring can be adjusted by the operator during a task. The proposed control strategies are implemented in a four-degree-of-freedom hydraulic manipulator referred to as HydCobot. Two experimental tasks for evaluating the accuracy, effort, and elapsed time are conducted to validate the effectiveness of the proposed methods. The results indicate that the proposed methods effectively enhance the performance of HRC.