Optimal Grasping Control System for Performing Precision Indirect Teaching of Robot Hand

The robot hand is connected to a manipulator and is widely used in the field of automated manufacturing processes, human cooperation processes and services. In addition, in an environment where there is a lot of danger because humans perform it directly, indirect teaching techniques that robots imitate human movements are used to perform tasks instead. However, if the torque of the robot hand is not taken into account during the teaching work, there is a risk of damage to the object or wasting current due to excessive torque. The torque of the robot hand described heremeans the current value of the motor. In this paper, we present a system that automatically adjusts the robot hand torque operating rate (TOR) by combining a contact sensing module and a fuzzy control system for precise indirect teaching. Torque operation rate refers to the control of the current value of the motor. After grasping an object, when the pressure sensing value of the contact sensing module is input to the fuzzy system, the TOR is controlled according to the fuzzy rule. This is expected to not only reduce damage to soft object, but also prevents current waste due to excessive hand operating torque.