The field of robotic chefs in the kitchen presents unique challenges that require bimanual manipulation to attain human-level dexterity. The incorporation of a second robot arm is widely believed to enhance the manipulation capabilities and open up new research and practical applications. However, identifying the overarching challenges and specific module-related obstacles that have impeded progress in leveraging this novelty for the benefit of humans remains a pressing issue. Addressing these challenges necessitates a comprehensive, interdisciplinary perspective that goes beyond solving isolated component-level problems. It is imperative for roboticists from academia and industry to collaborate, jointly reviewing, re-investigating, reshaping, and driving advancements in this research domain (ICRA2024 theme “CONNECT+”). To guide our discussions, we have formulated the following question as a focal point for the workshop:
What are the key challenges in bimanual manipulation for robotic chefs in the kitchen?
How can interdisciplinary collaboration enable innovative solutions, overcoming these obstacles to advance the research and practical implementation of robotic chefs?
How to transfer bimanual manipulation skills from human expert chefs to robots?
How to benchmark the robot’s cooking abilities?
What kind of sensor modalities are useful for robotic kitchen manipulation tasks?
How to best exploit the roles of manipulators for different functions in kitchen manipulation tasks, sensing, grasping, manipulation, exploration, etc?
How could computing tools, such as neural networks (including LLM), be a game changer for the dual/multi-arm coordination and manipulation for robotic kitchen tasks?
How to model the cooperation between human and robot chiefs so that the task can be implemented in an optimal way by taking into consideration the human preferences?