Motion Generation for Food Topping Challenge 2024: Serving Salmon Roe Bowl and Picking Fried Chicken

TL;DR

The paper tackles robotic food handling by combining four-channel bilateral control with motion-copying and imitation learning to achieve delicate, adaptive manipulation. It demonstrates two tasks from the ICRA 2024 Food Topping Challenge: fast, reproducible serving of salmon roe on rice using a motion-copying system, and high-count fried-chicken picking through bilateral-control-based imitation learning. The results show the method can deliver fast, stable food placement and high task success under dynamic conditions, though grasp failures and recovery remain challenging. The work highlights a practical pathway for deploying human-informed, contact-rich manipulation in automated food environments with potential gains in throughput and reliability.

Abstract

Although robots have been introduced in many industries, food production robots are yet to be widely employed because the food industry requires not only delicate movements to handle food but also complex movements that adapt to the environment. Force control is important for handling delicate objects such as food. In addition, achieving complex movements is possible by making robot motions based on human teachings. Four-channel bilateral control is proposed, which enables the simultaneous teaching of position and force information. Moreover, methods have been developed to reproduce motions obtained through human teachings and generate adaptive motions using learning. We demonstrated the effectiveness of these methods for food handling tasks in the Food Topping Challenge at the 2024 IEEE International Conference on Robotics and Automation (ICRA 2024). For the task of serving salmon roe on rice, we achieved the best performance because of the high reproducibility and quick motion of the proposed method. Further, for the task of picking fried chicken, we successfully picked the most pieces of fried chicken among all participating teams. This paper describes the implementation and performance of these methods.

Figures (12)

• Figure 1: Items for experiments.
• Figure 2: Foodly TypeR is a humanoid collaborative robot manufactured by RT Corporation. The left hand is the default hand, and the right hand is the cross hand used in this experiment. Both hands were replaced in the experiment. Sciurus17 is a research-use upper-body humanoid robot manufactured by RT Corporation.
• Figure 3: Block diagram of four-channel bilateral control.
• Figure 4: Block diagram of controller.
• Figure 5: Positioning for teaching movements.