IEEE ICRA 2021 Fullday Workshop (Online Event)
Bridging the Gap between Data-driven and Analytical Physics-based Grasping and Manipulation II
May 31, 2021
Abstract
Considerable progress in grasping and manipulation has been achieved using approaches that extract complex behaviors from data. Yet, data-driven approaches are mostly assessed empirically and not necessarily complying with physical and dynamical constraints compared to analytical approaches where these constraints can be modeled manually. Besides, application of black-box learning models often results in limited success due to large data requirements, incompetence in yielding physically consistent results, and lack of generalizability to novel cases. Meanwhile, physics-based approaches have also been improved dealing with uncertainty. Yet, simplifying assumptions on, e.g. contact and friction model, stationary environment, are often needed resulting in models that cannot account for variations arising when contact models are rich or environments are unstructured and dynamically change. As neither a learning-based nor an analytic approach can be considered sufficient for complex manipulation tasks with high dimensional state spaces, a continuum between mechanistic and learning models is indispensable, where both domain-specific knowledge and data are integrated synergistically. In contrast to practices based on simple forms of feature engineering, heuristics, and constraints, this workshop is focused on exploring a deeper coupling of learning-based methods with physics and discussing benefits of analytical and data-driven approaches in grasping and manipulation applications.
The goal of this event is to bring together researchers with different approaches to grasping and manipulation - both from the classical analytical modeling-based background and from the purely data-driven direction
Topics
The workshop will focus on the following areas, including but not limited to:
Learning and analytic approaches dealing with the uncertainty or unobservability in sensing and actuation duringgrasping and manipulation process
Simulation to reality transfer
Modeling, representation and integration of sensing modalities for grasp and manipulation tasks, e.g. proprio-ceptive, visual, force/torque, tactile, proximity sensing
Grasping of known, partially known or novel objects
New quality measures for grasping under uncertainty
Learning-based approaches for grasp planning and manipulation: e.g. model-based, model-free, data-efficient,multi-task, transfer, meta learning, reinforcement learning, learning from demonstration
Analytic and hybrid approaches for grasping and manipulation
Integration of data-driven with physics-based models for grasping and manipulation
How to model complex (object/hand/scene) interaction dynamics for grasp and manipulation tasks
Integrating learning and control for grasping and manipulation
Generalization and scalability of approaches to a variety of hands and objects
Approaches addressing deformable/flexible object manipulation, dexterous grasping and manipulation, in-handmanipulation, bi-manual manipulation, mobile manipulation (e.g. legged, wheeled, aerial, underwater manipu-lation)