About this Research Topic
Physical interaction of a robot with food items is present in a number of robotic applications (meal preparation, cooking, serving, feeding, etc.) spanning a wide variety of domains from food packaging to the service and hospitality industry, as well as in assistive applications. However, given the wide variety of food items in the real world and their widely varying physical properties, robotic food manipulation in the wild remains a huge challenge that encompasses a full stack of research problems ranging from perception, motion planning, control, and learning, to human-robot interaction, multimodal sensing, and mechanism design.
This Research Topic focuses on algorithms and technologies towards food manipulation. Some of the problems this Research Topic aims to tackle include:
• How to perceive and manipulate hard-to-model deformable items: Recent advances in multimodal sensing, perception, and learning have allowed robots to manipulate food items for applications such as cutting, feeding, cooking, etc. While this is useful, many challenges remain such as manipulating novel food items that have a wide variety of textures, compliance, and related physical properties.
• How to ensure safe, trustworthy, and effective handovers between robots and humans: Many applications such as serving and feeding would require a robot to handover food items to humans. While the robot-human handover literature is rich, it is still an open question on how to execute these handovers effectively in individual as well as group settings, while maintaining trust and safety in scenarios with evolving interpersonal dynamics.
• How to design mechanisms and systems that can enable robotic food manipulation in the wild: Leveraging intelligent mechanism design and soft robotics for robotic hands or grippers as well as full-stack systems combining sensing, perception, planning, control, and learning, it remains to be seen how all these technologies can come together towards solving a food manipulation application (meal preparation, cooking, serving, feeding, etc.) in unstructured environments.
Solving the above problems may require contributions from a diverse and interdisciplinary set of areas. This Research Topic invites both theoretical and experimental contributions in areas related to or towards robotic food manipulation, including, but not limited to:
Perception, Planning, and Learning
• Planning and reasoning during food manipulation (e.g. in the cooking application, it could be towards recipe and dish creation, cooking with limited resources, etc.)
• Robot learning for food manipulation (learning food dynamics and relevant manipulation skills from demonstrations and/or practice)
• Computer vision for food manipulation (finding ingredients, parts recognition of food, amount/quantity estimation of ingredients/seasonings, cooking/preparation/feeding state recognition)
• Non-visual perception in food manipulation (tactile, taste, smell, mechanical properties, etc.)
• Monitoring food during the cooking process (raw or overcooked? right shape?)
Robotic Systems, Technologies, and Hardware
• Robotic systems for food manipulation (cooking robots, meal preparation robots, meal serving robots, feeding robots, food material picking and placing, packaging, etc.)
• Robotic hands for food manipulation (underactuated hands, jamming hands, soft grippers, origami-structured hands)
• Sensors (RGB-D cameras, thermal cameras, tactile sensing, instrumented tools)
• Modeling and Simulation of food and food processes
• Food printing (3D printing)
• Deformable object manipulation (soft materials, thin materials, fragile materials)
• Manipulation with tools (knives, forks, spoons, tongs)
• Food manipulation skills (pouring, scooping, mixing, cutting, transporting, etc.)
• Knowledge base for food manipulation (recipes, cooking procedure, the physical and chemical process of ingredients, flavor, nutrition, health facts)
Social and/or Physical Human-Robot Interaction for user-centric food manipulation applications
• Human-robot interaction and collaboration in dyads and groups
• Safety in Robot-human handovers
• Trust and technology acceptance during robot-human handovers
• Human-in-the-loop food manipulation
• Modeling human intent for food manipulation
• User interfaces for communicating human intent
Keywords: Food manipulation, Deformable object manipulation, Human-robot interaction, Mechanism design, Robotic systems
Important Note: All contributions to this Research Topic must be within the scope of the section and journal to which they are submitted, as defined in their mission statements. Frontiers reserves the right to guide an out-of-scope manuscript to a more suitable section or journal at any stage of peer review.