主催: 一般社団法人 日本機械学会
会議名: ロボティクス・メカトロニクス 講演会2020
開催日: 2020/05/27 - 2020/05/30
In the manufacturing industries, improvement of productivity and stabilization of product quality are required, and automation of work by industrial robots is progressing. On the other hand, it is difficult for picking and kitting task to be automated, and workers perform it. Against this background, a support system for efficient picking, such as Digital Picking System (DPS), has been developed while preventing picking mistakes. However, it is considered that the work efficiency of the system currently used is reduced by the operation of confirming the picking task. Therefore, the authors proposed the Confirmation Free Picking System (CFPick), which recognizes the picking task automatically. In this paper, we report the result of task recognition of worker’s picking task using machine learning.