Introduction
Dexterous manipulation is a grand challenge in robotics and a necessary step in performing many manipulation tasks such as ones involving tool use. Today, many different approaches are being explored to tackle this problem: underactuated soft hands, fully actuated rigid hands, model-based control, reinforcement learning, imitation learning, experiments with many different tactile sensors, simulators for contact-based interactions, and many others. Furthermore, there is little convergence on dexterous manipulation tasks being solved by different research groups. Due to the diversity of approaches and lack of dialogue between different sub-communities, quantifying progress and identifying challenges and opportunities is a formidable issue. We believe that big jumps in dexterous manipulation will result from interdisciplinary work spanning the subfields of control, hardware, sensor and simulation design. We aim to bring together researchers from these diverse communities to encourage dialogue, identify avenues for collaboration and understand the status, challenges, and opportunities for interdisciplinary work. For seasoned researchers, this workshop will be an avenue for exchanging ideas with a diverse community. For students and early career researchers, this workshop will be an opportunity to understand the landscape and future directions of dexterous manipulation.
- Optimization/Learning/Planning for dexterous manipulation
- Compliant control in dexterous manipulation
- Teleoperation for dexterous manipulation
- Novel manipulator mechanisms for dexterous manipulation
- Low-cost hardware design for dexterous manipulation
- Novel sensor designs enabling new dexterous manipulation capabilities
- Tactile Sensing for dexterous manipulation
- Simulation for contact-rich manipulation
- ...
This interdisciplinary workshop will convene researchers across disciplines to assess the status quo in dexterous manipulation research and illuminate potential future directions worthy of pursuit.
Invited Speakers
listed alphabetically
Call for Papers
We invite submission to the Ingredients of Dexterous Manipulation workshop, hosted at ICRA 2024.
Paper topics
A non-exhaustive list of relevant topics:
In this workshop, we aim to bring together machine learning and robotics researchers who work at the intersection of these fields. We invite researchers to submit work in the following or related areas (non-exhaustive list):
- Planning and Control in Dexterous Manipulation:
- What methods can be used to model the contact between a dexterous hand and an object to aid in control and planning?
- What strategies can be used to manage high DoF systems with control methods?
- Designing Hardware for Dexterous Manipulation:
- Is it possible to construct a cost-effective manipulator for dexterous manipulation?
- Does the soft hand model contribute to better manipulation tasks, and how can it be modeled?
- Collecting Dexterous Manipulation Data:
- Can we collect general-purpose human hand data for dexterous manipulation using any professional equipment? What differences exist between human and robot hand data?
- What improvements can be made to current data collection methods, like teleoperation, to enable large-scale data gathering?
- Tactile Sensing for Manipulation:
- How can tactile data aid robots in better task execution and environmental perception?
- What type of tactile information is most effective for dexterous robot hands, and can it make up for visual perception limitations?
- Learning for Dexterous Manipulation:
- Can we expect a comprehensive and generalized model for most everyday dexterous manipulation tasks, or will each task require its own specialized model?
- How can learning-based policies manage dynamic tasks that demand high-frequency control and detailed dynamic models?
- Any additional related topics not already covered in the above list 😄
Submission Guidelines (TBD)
