Current-Based Direct Teaching for Industrial Manipulator

Abstract

In this paper, a novel human direct teaching method of industrial manipulator with large mass and friction based on internal current sensors installed in joints is proposed. Dynamic model for an industrial robot is analyzed in detail, where the gravity torque and friction torque are the main items that need to be considered. Friction torque is influenced by joint speed and temperature. A method to identify joint friction parameters rapidly is proposed, and experiments are carried out to acquire friction characteristics under different temperatures. Then, the implementation method and improvement measures of direct human teaching are introduced in consideration of robot gravity torque and friction torque of joints. Finally, validation tests are conducted to verify the feasibility of the direct teaching method. Experiment result shows that the method based on internal current sensors is capable of realizing direct teaching flexibly and easily.