Parts-Picking based on Visual and Force Information in Disordered Environment.

This paper proposes approaches to reduce the processing load on a vision system and to detect accidents using a force/torque sensor in real time. For the former approach, the concepts of the parts isolation factor and robot manipulability at the grasping position are introduced to formulate a parts-picking strategy for executing the task while maintaining the initial environment. The picking operation is executed in the order of the highest isolation factor and manipulability. An accident recovery method in which force and torque are monitored during operation is also presented. Successful parts-picking by robot systems implementing these approaches was confirmed by simulation and by actual robot operations.