News

• Best paper awards have been announced (see below). Congratulations to all the awardees and authors.

• The list of accepted papers, including manuscripts and videos, is now available. See you at the poster session.

• The workshop will take place in room 330A on October 5th, starting at 8.30AM

• We are pleased to welcome our Gold Sponsor: Agile Robots AG

About

The 1st International Workshop “Learning Meets Model-based Methods for Manipulation and Grasping” will take place as part of IEEE/RSJ IROS 2023 on October 5th, 2023 in Detroit, USA.

This event upholds the IEEE RAS code of conduct to provide a safe, productive, and welcoming environment for all.

Awards

Best Paper Award

AnyGrasp: A Library for Human-level Grasp Perception in Cluttered and Dynamic Environments
Hao-Shu Fang (SJTU)*; Chenxi Wang (Shanghai Jiao Tong University); Hongjie Fang (Shanghai Jiao Tong University); Minghao Gou (Shanghai Jiao Tong University); Jirong Liu (Shanghai JiaoTong University); Hengxu Yan (Shanghai Jiao Tong University); Wenhai Liu (Shanghai Jiao Tong University); Yichen Xie (shanghai jiao tong university); Cewu Lu (Shanghai Jiao Tong University)
Paper

Ex-aequo with

Dynamic-Resolution Model Learning for Object Pile Manipulation
Yixuan Wang (University of Illinois at Urbana-Champaign); Yunzhu Li (Stanford University & University of Illinois at Urbana-Champaign)*; Katherine Driggs-Campbell (University of Illinois at Urbana-Champaign); Li Fei-Fei (Stanford University); Jiajun Wu (Stanford University)
Paper

Agile Robots Best Student Paper Award

Learning Plan-Satisficing Motion Policies from Demonstrations
Yanwei Wang (MIT)*; Nadia Figueroa (UPenn); Shen Li (MIT); Ankit Shah (Brown University); Julie Shah (MIT)
Paper

Runner-up Best Student Paper Award

Neural Field Movement Primitives for Joint Modelling of Scenes and Motions
Ahmet Tekden (Chalmers University of Technology)*; Marc Deisenroth (University College London); Yasemin Bekiroglu (Chalmers)
Paper Video

Abstract

Building robots capable of dexterous interaction with objects to carry out fine manipulation tasks has always been a grand challenge in robotics. The non-smooth, brittle nature of manipulator-object mechanics, together with perceptual uncertainty, easily violate the assumptions of early planning and control methods. Furthermore, accurate physical modeling of complex or non-rigid mechanical systems requires large amounts of computations, which is incompatible with real-time control.

Such challenges led researchers to develop a wide range of approaches, from adaptive control tailored to the (potentially changing) properties of the object at hand, to advanced perception to tackle measurement uncertainty. Machine learning also contributed by providing actionable representations of complex geometries and visual appearance, and by encoding hard-to-model expert demonstrations to reduce the cost of trial-and-error. In turn, these informed the development of novel robot control methods enabling more robust and dexterous skills. At the same time, the employment of mechanical models proved effective for enforcing structural constraints in robot control systems (including learning-based ones), thus improving safety and guiding exploration.

However, there are still many open challenges that need to be addressed to achieve long-horizon robotic manipulation and sidestep the computational burden of accurate simulation of contact-rich scenarios. The ambition of this workshop is to provide a comprehensive overview of the broad and scattered state of the art in robot manipulation and grasping, spanning model-based and learning-based approaches. Talks and interactive sessions will enable a deeper understanding of current approaches in different use cases, while stimulating the development of new methods.

Gold Sponsor

Acknowledgments