Hybrid force/position control in workspace of robotic manipulator in uncertain environments based on adaptive fuzzy control

Abstract

In this study, a novel hybrid force/position controller in workspace of robotic manipulator based on adaptive fuzzy control is developed to improve the controlling performance of force and position in contact with uncertain environments. The dynamic model of the robotic manipulator in joint space is converted to a dynamic model in workspace, which is consisted of the model of position-controlled subsystem and model of force-controlled subsystem. Furthermore, in the position-controlled subsystem, an adaptive fuzzy computed torque control is proposed by considering adaptive fuzzy control and conventional computed torque control (AFCTC), for compensating deviations caused by the presence of structured uncertainty and unstructured uncertainty. In the force-controlled subsystem, a fuzzy proportional integral control method (FPI) is proposed by fuzzy logic and conventional proportional integral method to improve the control performance. The asymptotic stability of the developed controller is proved by Lyapunov theorem. The simulation results show the preferable performance of the proposed controller (AFCTC-FPI) by comparison with adaptive fuzzy sliding mode control (AFSMC), adaptive network-based fuzzy inference system with proportion differential and integral (ANFIS-PD+I), computed torque control-proportion integral (CTC-PI), fuzzy proportional integral derivative (FPID), and proportional integral derivative (PID) in uncertain environments. Furthermore, in the experimental study, average position error with AFCTC-FPI is decreased by 87.14 % than that with PID, meanwhile, range of interaction force with AFCTC-FPI is reduced by 70.31% than that with PID. In summary, the experimental results also show the superior control accuracy of the proposed controller in real environment.