RSS 2015 Workshop
Bridging the Gap between Data-driven and Analytical Physics-based Grasping and Manipulation
July 16, Rome
Abstract
In the last two decades, probabilistic data-driven approaches to manipulation have celebrated extensive successes due to their ability to extract complex behaviors from collected data. Generative probabilistic approaches furthermore have the advantage that they can synthesize grasps based on previously seen data. However these approaches can often only be verified empirically and do not necessarily abide by physical and dynamical constraints which can be modeled manually in an analytical approach. At the same time, these classical physics-based approaches to manipulation have advanced, and have started to incorporate a notion of uncertainty in their analysis. However simplifying assumptions e.g. on the contact and friction model, often have to be made. The combination of physics-based constraints with probabilistic reasoning however remains one of the key challenges in this field. This workshop focuses on bridging this gap, as well as on discussing the respective benefits of analytical and probabilistic approaches in various application domains.
Topics
The workshop will focus on the following areas, including but not limited to:
- Grasping of known, partially known or novel objects
- New quality measures for grasping under uncertainty
- Representations incorporating uncertainty for grasping and manipulation
- Grasping and Manipulation based on Multi-modal Data
- Learning-based approaches for grasp planning
- Integration of data or model learned from data with the physics in grasping and manipulation
- How to reduce the complexity of the physics
- Approaches dealing with the uncertainty or unobservability in grasping process
- Approaches based on physics and learning
- Adapting results to different mechanical hands
Schedule
Workshop Agenda
The aim is to bring together researchers with different approaches to grasping and manipulation, both from the classical analytical/physics modeling based background and from the purely data-driven direction. The proposed speakers have been selected to cover a wide range of probabilistic and analytical grasp synthesis methods and to include experts in robotics, machine learning, and control. Invited speakers will get a 30 minute slot to present their latest work relevant to the workshop theme, also including the limits and opportunities that they see with these approaches and how they can be combined and feature in their own work. We will organize an interactive poster session, allowing more contributors to present their work and have detailed discussions with the workshop attendees. Contributed papers will be published on the website and invited to present a poster. We will have a panel discussion at the end where participants can discuss open questions on the state of the art in grasp and manipulation planning and the proposed directions of new research efforts that will be raised during the talks by the speakers. We hope to include many contributors and have fruitful discussions between invited speakers, active contributors, and the audience.
Call for Papers
We welcome the submission of two page extended abstracts describing new or ongoing work. The best two submissions will be selected for oral presentations during the workshop. Final instructions for poster presentations and talks will be available on the workshop website after decision notifications have been made. Accepted presenters will have the option of submitting a full length six-page paper. All contributed papers will be accessible on the workshop website. Submissions should be in .pdf format using the RSS paper template (LaTeX template, Word template ). Please send submissions to one of the organizers with the subject line "RSS 2015 Workshop Submission”. For any questions or clarifications, please contact the organizers.
Important Dates:
Abstract submission deadline: June 26, 2015
Acceptance notification: July 3, 2015
Final materials due: July 9, 2015
Workshop date: July 16, 2015
Organizers
Florian T. Pokorny (UC Berkeley), ftpokorny [at] berkeley.edu
Yu Sun (USF), yusun [at] cse.usf.edu
Carl Henrik Ek (KTH), chek [at] csc.kth.se
Yasemin Bekiroglu (KTH), yaseminb [at] csc.kth.se