In:
ACM Transactions on Graphics, Association for Computing Machinery (ACM), Vol. 39, No. 4 ( 2020-08-31)
Abstract:
We address the longstanding challenge of producing flexible, realistic humanoid character controllers that can perform diverse whole-body tasks involving object interactions. This challenge is central to a variety of fields, from graphics and animation to robotics and motor neuroscience. Our physics-based environment uses realistic actuation and first-person perception - including touch sensors and egocentric vision - with a view to producing active-sensing behaviors (e.g. gaze direction), transferability to real robots, and comparisons to the biology. We develop an integrated neural-network based approach consisting of a motor primitive module, human demonstrations, and an instructed reinforcement learning regime with curricula and task variations. We demonstrate the utility of our approach for several tasks, including goal-conditioned box carrying and ball catching, and we characterize its behavioral robustness. The resulting controllers can be deployed in real-time on a standard PC. 1
Type of Medium:
Online Resource
ISSN:
0730-0301
,
1557-7368
DOI:
10.1145/3386569.3392474
Language:
English
Publisher:
Association for Computing Machinery (ACM)
Publication Date:
2020
detail.hit.zdb_id:
2006336-2
detail.hit.zdb_id:
625686-7
Permalink