A Robotic System Capable of Recognition, Grasping, and Suction for Dishwashing Automation

Abstract

In the food services industry, automation is increasingly demanded due to labor shortages caused by the super-aged society in Japan. There are several challenges upon introducing robotic systems into the food services industry, such as the lack of effective robotic end-effectors for handling various tablewares, recognition of tablewares in complex environment, and high-speed motion to ensure efficiency. In this paper, a robotic system was developed to perform the taking-out operation of various tablewares from a commercial dishwasher. The system consists of a 6-DOF robotic manipulator, a robotic hand capable of grasping and suction operations, and a 3D camera for tableware recognition. Machine learning approach was used to classify the tableware categories and obtain the grasping positions and postures of the tablewares. Experiments were conducted on taking-out several tablewares from a dishwasher rack and results validated the proposed robotic system for potential applications of automating dishwashing tasks in the food services industry.