Learning to Transfer In-Hand Manipulations Using a Greedy Shape Curriculum
| dc.contributor.author | Zhang, Yunbo | en_US |
| dc.contributor.author | Clegg, Alexander | en_US |
| dc.contributor.author | Ha, Sehoon | en_US |
| dc.contributor.author | Turk, Greg | en_US |
| dc.contributor.author | Ye, Yuting | en_US |
| dc.contributor.editor | Myszkowski, Karol | en_US |
| dc.contributor.editor | Niessner, Matthias | en_US |
| dc.date.accessioned | 2023-05-03T06:09:23Z | |
| dc.date.available | 2023-05-03T06:09:23Z | |
| dc.date.issued | 2023 | |
| dc.description.abstract | In-hand object manipulation is challenging to simulate due to complex contact dynamics, non-repetitive finger gaits, and the need to indirectly control unactuated objects. Further adapting a successful manipulation skill to new objects with different shapes and physical properties is a similarly challenging problem. In this work, we show that natural and robust in-hand manipulation of simple objects in a dynamic simulation can be learned from a high quality motion capture example via deep reinforcement learning with careful designs of the imitation learning problem. We apply our approach on both single-handed and two-handed dexterous manipulations of diverse object shapes and motions. We then demonstrate further adaptation of the example motion to a more complex shape through curriculum learning on intermediate shapes morphed between the source and target object. While a naive curriculum of progressive morphs often falls short, we propose a simple greedy curriculum search algorithm that can successfully apply to a range of objects such as a teapot, bunny, bottle, train, and elephant. | en_US |
| dc.description.number | 2 | |
| dc.description.sectionheaders | Human Object Interaction | |
| dc.description.seriesinformation | Computer Graphics Forum | |
| dc.description.volume | 42 | |
| dc.identifier.doi | 10.1111/cgf.14741 | |
| dc.identifier.issn | 1467-8659 | |
| dc.identifier.pages | 25-36 | |
| dc.identifier.pages | 12 pages | |
| dc.identifier.uri | https://doi.org/10.1111/cgf.14741 | |
| dc.identifier.uri | https://diglib.eg.org:443/handle/10.1111/cgf14741 | |
| dc.publisher | The Eurographics Association and John Wiley & Sons Ltd. | en_US |
| dc.subject | CCS Concepts: Computing methodologies -> Physical simulation; Motion capture; Reinforcement learning; Learning from demonstrations | |
| dc.subject | Computing methodologies | |
| dc.subject | Physical simulation | |
| dc.subject | Motion capture | |
| dc.subject | Reinforcement learning | |
| dc.subject | Learning from demonstrations | |
| dc.title | Learning to Transfer In-Hand Manipulations Using a Greedy Shape Curriculum | en_US |