Position and force hybrid control of robotic manipulator by neuralnetwork (adaptive control of 2 DOF manipulators)

摘要

A position/force hybrid control of a robotic manipulator based onna neural network model is proposed with consideration of the dynamics ofnobjects and the orientations of the robotic manipulator. This proposednsystem consists of a standard PID (proportional plus integral plusnderivative) controller, the gains of which are augmented and adjustedndepending on objects and orientations of manipulators through a processnof learning. The proposed method shows better performance than thenconventional PID controller, yielding a wider range of applications. Itnis shown that the proposed controller is applicable to cases ofnposition/force hybrid control of multi-degree-of-freedom manipulators.nSimulations and experiments were carried out for the case ofntwo-degree-of-freedom robotic manipulators