Learning of Working Space based on Joint Overloads for a Multi-DOF Manipulator

Abstract

This paper describes a novel method of probabilistic self-modeling based on learning of operating space from exploratory actions of a multi-DOF manipulator, which is designed for mounting it onto the wheelchair. The developed anthropomorphic multi-DOF manipulator is able to learn both of the operating range in each joint and the probabilistic operating space based on Gaussian Mixture Model and Variational Bayesian learning algorithm. We introduce an acquisition method of the operating space by using the historical data of irregular overload, which is detected by using analogue current signals measured by solely internal sensor of joint motors. In addition, online behavior learning with a simple probabilistic path planning is also presented based on the obtained probabilistic operating space. We will conduct several experiments with a real multi-DOF manipulator arm. After the basic characteristics of the obtained operating space are shown, the performance of interaction with different situations such as different load given to the arm and obstacles placed in the surrounding environment will also be demonstrated.