Abstract
In this paper, a force control for manipulator which utilizes a neural network model is proposed with consideration of dynamics of objects. This proposed system consists of the standard PID controller and the perceptron-like neural model, which tunes a set of controller's parameters via a process of learning. The neural network model has not been applied to force control problems yet, but this proposed method is shown to be applicable to the force/compliance control problems. Stability of this system and wider applicability are verified by simulation studies.
