Human-robot cooperation for heavy object manipulation based on sensing pneumatic pressure in air cushions

Abstract

The objective of this study is to develop a novel method for human-robot cooperation for heavy object manipulation based on sensing pneumatic pressure in air cushions. Forearm-support type lift assist system (FOLAS), a human-cooperative robot we developed, has air cushions that works as a buffer between the machine and the user's forearm. The operating intention of the user can be indirectly detected by measuring the variation of the pneumatic pressure inside the air cushions. Force control to compensate the weight of a lifted object was adopted for the vertical operation of the FOLAS on the assumption that force applied to an air cushion has proportional relationship to pneumatic pressure in the air cushion. Position control by using overlap length between movable cushion and fixed cushion was adopted for the horizontal operation, and the overlap length was estimated by pneumatic pressure ratio in the air cushions. An experiment for the operational performance showed the FOLAS could be operated forward and backward according to the intention of a user continuously supporting his/her forearm and a heavy object.