(RObotic MAnipulation of Deformable Objects)
Workshop at IROS 2020
(IEEE/RSJ International Conference on Intelligent Robots and Systems)
The Caesars Forum Convention Center, Las Vegas, USA, October 29, 2020
Novel solutions for the robotic manipulation of deformable objects can enable new real-world tasks in areas of strategic interest. These include the manufacturing industry, agriculture, or medicine. However, deformable objects are complex to model and require the manipulation system to have high adaptability and the ability to operate under uncertainty. These are important challenges and, although exciting progress has been made in the field over the past years, there remain many open technical questions. In particular, the workshop targets the following topics of interest:
1) Planning and control for grasping and manipulation of deformable objects.
Grasping and manipulation of deformable objects present great challenges because non-rigidity increases the uncertainty and the complexity of modelling. Current research efforts are based on prior identification of the object’s deformation properties or use a sensor-based definition of the task with online adaptation. However, the performance of existing approaches is still limited by the challenges mentioned above. Another major issue is the scarcity of general, as opposed to ad-hoc, frameworks for deformable object grasping and manipulation.
2) Multi-modal perception for manipulation of deformable objects.
Multi-modal sensing approaches are inspired by the human perception based on exploiting both touch and vision senses. They are almost a necessity when a robot has to perform complex manipulation tasks such as controlling the shape of a deformable object or ensuring a stable grasp. Both vision and tactile sensors become more challenging to use effectively when the object to be sensed is deformable. New trends are focused on combining or fusing these two sensing modalities, as well as the proprioceptive identification from the robot.
3) Machine learning for manipulation of deformable objects.
Deep learning from vision and touch data can be used for recognizing physical features and surface reconstruction of manipulated objects to determine how the robot should grasp the object or to predict its movements. Also, deep reinforcement learning can be used, e.g., to control a robotic hand to continuously update the grasp strategy with a wide variety of objects. Machine learning is undoubtedly a very promising tool in the highly uncertain and complex scenarios of deformable object manipulation.
4) Multi-robot and human-robot interactions in the manipulation of deformable objects.
The use of multiple robots (as opposed to a single one) allows transporting large deformable objects or controlling their shape in a more refined way. Perception, planning and control for such complex tasks should exploit both (i) multi-robot coordination and optimization frameworks and (ii) the deformation properties of the object. The integration of the two aspects in a synergistic manner remains an underexplored topic. Moreover in prominent cases of practical interest a robot has to manipulate a deformable object while interacting with humans. This also calls for novel approaches to take deformability into account in the definition and control of the interaction.
Topics of interest
The topics of interest include, but are not limited to:
- Grasping for deformable objects
- Dexterous manipulation for deformable objects
- Manipulation planning for deformable objects
- Shape control of deformable objects
- Visual sensing for manipulation of deformable objects
- Tactile sensing for manipulation of deformable objects
- Multimodal perception for manipulation of deformable objects
- Machine learning for manipulation of deformable objects
- Multi-robot manipulation of deformable objects
- Human-robot manipulation of deformable objects
Call for contributions
14 August, 2020: Paper submissions
31 August, 2020: Acceptance notification
14 September, 2020: Camera ready
29 October, 2020: Workshop date
Instructions for Authors
We solicit contributions of two types:
– Short papers (2-4 pages): Concise descriptions of preliminary results and ongoing work.
– Regular papers (4-8 pages): More detailed descriptions of ongoing work or of more mature developments.
Papers should be in PDF format and prepared using the standard IROS template. They will be reviewed and their acceptance will be decided on the basis of their scientific merit and relevance to the workshop’s topics of interest. The accepted papers will be made available on the workshop’s website. The accepted papers will be presented as posters at the workshop, with the possibility of oral presentation for selected papers.
Papers should be submitted by email to email@example.com. Please identify clearly the type of submission (“Short” or “Regular”) in the subject of the email. An acknowledgement of receipt will be sent shortly after submission.
We will consider the publication of an independently peer-reviewed journal special issue after the workshop. Authors of papers presented at the workshop would be invited to submit extended versions to the special issue.
Note that we are ready to hold the workshop virtually if required due to the COVID-19 pandemic.
Alberto Rodríguez, MIT
David Navarro-Alarcón, The Chinese University of Hong Kong
Danica Kragic, KTH Royal Institute of Technology
Berk Calli, Worcester Polytechnic Institute
Raúl Suárez, Universitat Politècnica de Catalunya
Kaspar Althoefer, Queen Mary University of London
Véronique Perdereau, Université Pierre et Marie Curie
Javier Alonso-Mora, Delft University of Technology
Zhanat Kappasov, Nazarbayev University
Miguel Aranda (SIGMA Clermont, Institut Pascal)
Juan Antonio Corrales (SIGMA Clermont, Institut Pascal)
Pablo Gil (Universidad de Alicante)
Gonzalo López-Nicolás (Universidad de Zaragoza)
Helder Araujo (Universidade de Coimbra)
Youcef Mezouar (SIGMA Clermont, Institut Pascal)
This workshop received support from the IEEE RAS Technical Committee on Robotic Hands, Grasping, and Manipulation and from the IEEE RAS Technical Committee on Multi-Robot Systems.
For any question regarding the workshop send an email to firstname.lastname@example.org