Action oriented self-modeling and motion planning for a humanoid robot

This paper describes a new method of self-modeling based on active acquisition of operating ranges for a humanoid robot. This approach takes insights from pain perception, which is regarded as a function of self-preservation in nature. The anthropomorphic humanoid robot determines the operating ranges of joint actuators and the workspace by its own active behavior. In addition, in order to protect the joint actuators, the robot is operated using an algorithm that detects irregular contact. We also demonstrate that the developed robot is robust against dynamical changes in the surrounding environment. For an arm having 5 degrees of freedom (DOFs), path planning is performed in the joint-angle space by utilizing the proposed method and Rapidly-exploring Random Trees (RRTs). We conduct several experiments in a real environment in order to verify the advantages of the proposed approach.