Bimanual Manipulation: Learning, Planning and Control
Objectives
Objectives
Bimanual manipulation (dual-arm manipulation) encompasses a large variety of research activities. Human-centered environments provide affordances for and require the use of dual-arm, two-handed, or bimanual, manipulations. The coordination between two arms greatly increases the dexterity and flexibility of robot manipulation and therefore open new directions for both research and applications. In bimanual manipulation, one arm is usually playing as a grasper while the other arm is playing as a manipulator to interact with the grasped object. The two arms can further switch role during the manipulation process according to the real-time sensory information and high-level task priority planning. The driving force behind this interest is the vision that many practical robot applications, such as assembly, assist, which generally require the use of two arms to work together to accomplish the task or to support human beings. Current robots designed to function in, and physically interact with, these environments have not been fully able to meet these requirements because standard bimanual control approaches have not accommodated the diverse, dynamic, and intricate coordination between two arms. With recent trend, bimanual manipulation technologies are demanding to bring robots even closer to human to accomplish more complicated but mandatory tasks and assist more daily life but challenging tasks, with more and more easily accessible and affordable robot hardware (humanoids, collaborative arms, legged systems) in the market. It is important for roboticists from the community to join the effort together to review and investigate the future of bimanual manipulation technologies. To address this issue, we need to answer the following questions.
Bimanual manipulation (dual-arm manipulation) encompasses a large variety of research activities. Human-centered environments provide affordances for and require the use of dual-arm, two-handed, or bimanual, manipulations. The coordination between two arms greatly increases the dexterity and flexibility of robot manipulation and therefore open new directions for both research and applications. In bimanual manipulation, one arm is usually playing as a grasper while the other arm is playing as a manipulator to interact with the grasped object. The two arms can further switch role during the manipulation process according to the real-time sensory information and high-level task priority planning. The driving force behind this interest is the vision that many practical robot applications, such as assembly, assist, which generally require the use of two arms to work together to accomplish the task or to support human beings. Current robots designed to function in, and physically interact with, these environments have not been fully able to meet these requirements because standard bimanual control approaches have not accommodated the diverse, dynamic, and intricate coordination between two arms. With recent trend, bimanual manipulation technologies are demanding to bring robots even closer to human to accomplish more complicated but mandatory tasks and assist more daily life but challenging tasks, with more and more easily accessible and affordable robot hardware (humanoids, collaborative arms, legged systems) in the market. It is important for roboticists from the community to join the effort together to review and investigate the future of bimanual manipulation technologies. To address this issue, we need to answer the following questions.
- How to coordinate motion of two arms in a synergistic way that takes task constraints into account?
- How to best exploit the roles of arms for different functions, sensing, grasping, manipulation, etc?
- How to optimize the combination of uni-manual and bi-manual manipulation behaviors？
- How to design the manipulation process in an intuitive way for a given task?
- How to transfer human bimanual manipulation skills to robots?
- What kind of sensor modalities are useful for bimanual manipulation?
- How to model the cooperation between human and robots so that the task can be implemented in an optimal way by taking into consideration the human preferences?
- How could the computing tools, such as neural networks (including deep learning), be a game changer for the dual/multi arm coordination and manipulation?
- How would all kinds of mobile platforms contribute to the efficiency of the bimanual manipulation?
Important Dates
Important Dates
Poster abstract submission deadline: 21 May 2021
Notification of acceptance: 23 May 2021
Video submission deadline: 27 May 2021
Workshop (virtual): 31 May 2021
Workshop (virtual): 31 May 2021
10:30-18:30 (at GMT+8 – Xi’an/Beijing)
10:30-18:30 (at GMT+8 – Xi’an/Beijing)
4:30-12:30 (at CET)
4:30-12:30 (at CET)
19:30 (on May 30th) - 3:30 (on May 31st, at PDT)
19:30 (on May 30th) - 3:30 (on May 31st, at PDT)
ZOOM Meeting:
ZOOM Meeting:
You can also use zoom meeting ID: 990 7485 8244 to login. Please note that a quick registration needed upon entering the zoom meeting room.
You can also use zoom meeting ID: 990 7485 8244 to login. Please note that a quick registration needed upon entering the zoom meeting room.
Topics of Interests
Topics of Interests
- Bio-inspired/human-inspired bimanual manipulation theories and studies
- Bimanual manipulation task coordination and planning
- Bimanual manipulation task sensing and learning
- Bimanual manipulation control for robot arms and hands
- Bimanual (loco-)manipulation with various mobile platforms, wheels, legged, etc.
- Robot mechanics and electronics for bimanual manipulation hardware setup
- System communication and optimization for effective bimanual manipulation
- Bimanual manipulation in different applications: manufacturing, healthcare, logistics, etc.
- Multi-arm manipulation beyond dual-arm bimanual manipulation
Organizers
Organizers
- Fei Chen, Assistant Professor, T-Stone Robotics Institute, The Chinese University of Hong Kong, Hong Kong SAR | Department of Advanced Robotics, Italian Institute of Technology, Italy. Email: f.chen@ieee.org (main contact)
- Sylvain Calinon, Senior Researcher, Idiap Research Institute, Switzerland. Email: sylvain.calinon@idiap.ch
- Huan Tan, Director, UBTECH North America Research and Development Center, US. Email: huan.tan@ubtrobot.com
- Yunhui Liu, Professor/Director, T-Stone Robotics Institute, The Chinese University of Hong Kong, Hong Kong SAR. Email: yhliu@cuhk.edu.hk
Support TC
Support TC
We are grateful for the support of the following organizations:
We are grateful for the support of the following organizations:
Mobile Manipulation
Robot Learning
Robotic Hands, Grasping and Manipulation
Collaborative Automation for Flexible Manufacturing
Algorithms for Planning and Control of Robot Motion