Tactile Sensing-Based Control Architecture in Multi-Fingered Arm for Object Manipulation

This report presents the development of tactile sensing-based control architecture in a multi-fingered humanoid robot arm for object manipulation tasks. With the aim to enhance the ability to recognize and manipulate object in humanoid robot, we developed a novel optical three-axis tactile sensor system mounted on fingertips of the humanoid robot fingers. This tactile sensor applies an optical waveguide transduction method, and capable of acquiring normal and shearing force. Trajectory generation based on kinematical solutions at the arm and fingers, together with control system structure and sensing principle of the tactile sensor system are presented. We proposed control algorithm based on tactile sensing in the robot control system. Object manipulation experiments are conducted using hard and soft objects. Experimental results revealed that the proposed control system enable the finger system to recognize low force interactions based on tactile sensing information to grasp the object surface and manipulate it without causing damage to the object and the sensor elements.