Hybrid force/position control of robotic arms manipulating in uncertain environments based on adaptive fuzzy sliding mode control

In this study, the hybrid force/position control of robotic manipulators operating in uncertain environments is addressed by integrating the fuzzy logic with conventional sliding mode control (SMC). After decomposition of the manipulator dynamics into position, force, and redundant joint subspaces, the universal approximation capability of fuzzy systems is employed to approximate the equivalent part of the control input constructed based on SMC concept. The robust part of the controller is estimated by an adaptive PI controller to compensate for deviations due to the presence of model uncertainties and perturbations. Furthermore, some adaptation laws are derived for updating the parameters online when some changes in the system dynamics are made. The proposed adaptive fuzzy sliding mode control (AFSMC) requires the minimum information about the manipulator dynamic structure and environment physical properties among the other hybrid force/position control methods presented so far. Indeed, it needs neither estimation of the dynamic model nor bounds of uncertainties in advance. The asymptotic stability of the proposed controller is also proved in the sense of Lyapunov theorem. The simulation results show the good performance of the proposed controller in coping with uncertainties. The proposed scheme is also compared with standard SMC methodology, and its superior robustness is shown in comparison with those methods which require an estimation of the plant mathematical model.