Tactile Sensing and Control in Humans and Robotic/Teleoperated Systems
STANFORD UNIV CA DEPT OF MECHANICAL ENGINEERING
Pagination or Media Count:
The most significant result of the StanfordHarvardUmea collaboration is a promising new approach to grasp force control in manipulation with robots and teleoperators. Experiments at Umea showed that human subjects control grasp forces and apply manipulation strategies using information from fast-acting FAI and FAII tactile afferents which signal events such as the onset of slipping between the fingers and a grasped object. An information exchange between Umea and Stanford led to experiments that demonstrated that the output of these FAI and FAII sensors is occasionally accompanied by small, but detectable, redistributions in the tangential contact forces between the fingertips of the human subjects and the objects that they were grasping. Dynamic tactile sensors, specifically the skin acceleration and the stress rate sensors, were developed at Stanford for robotic and teleoperated hands to provide information similar to the human PA sensors. Triggering off the microslip information provided by these sensors, a manipulators force control algorithm can then quickly adjust the grasp force so that no gross sliding occurs. Tactile sensing, robotics, physiology, dextrous manipulation, teleoperation.
- Information Science
- Computer Programming and Software