A generalized adaptive stiffness control scheme for robot manipulators with bounded inputs

Abstract

In general, indirect force control schemes (stiffness, impedance, etc.) assume that robot actuators can provide any torque value to achieve the goal of interaction control. This study attempts to regulate robot–environment interaction by generating bounded control signals and to avoid accurate knowledge of the parameters associated with gravitational effects and the stiffness of the environment. To achieve this aim, a generalized and saturating adaptive stiffness control scheme in task‐space is proposed. For the purpose of this work, the interaction or contact between the end‐effector of a robot manipulator and the environment is modeled as a vector of bounded spring‐like forces. The proposed control approach has a proportional‐derivative structure with static model‐based compensation of gravitational and interaction forces, which it achieves by including a regressor‐based adaptive term. As a theoretical basis to support the proposal, Lyapunov's stability analysis of the closed‐loop equilibrium vector is presented. Finally, the suitability of the proposed stiffness control scheme for interaction tasks is verified through simulations and experimental tests by using three‐degree‐of‐freedom robotic arms.