Scaling simulation-to-real transfer by learning composable robot skills

Ryan Julian
Ryan Julian
Eric Heiden
Eric Heiden
Zhanpeng He
Zhanpeng He
Hejia Zhang
Hejia Zhang

International Symposium on Experimental Robotics, Volume abs/1809.10253, 2018.

Cited by: 4|Bibtex|Views56
EI
Other Links: dblp.uni-trier.de|academic.microsoft.com|arxiv.org

Abstract:

We present a novel solution to the problem of simulation-to-real transfer, which builds on recent advances in robot decomposition. Rather than focusing on minimizing the simulation-reality gap, we learn a set of diverse policies that are parameterized in a way that makes them easily reusable. This diversity and parameterization of low-le...More

Code:

Data:

Your rating :
0

 

Tags
Comments