VR-Goggles for Robots: Real-to-sim Domain Adaptation for Visual Control release_bqae25kvurctfndefni5ufm5km

by Jingwei Zhang, Lei Tai, Peng Yun, Yufeng Xiong, Ming Liu, Joschka Boedecker, Wolfram Burgard

Released as a article .

2018  

Abstract

In this paper, we deal with the reality gap from a novel perspective, targeting transferring Deep Reinforcement Learning (DRL) policies learned in simulated environments to the real-world domain for visual control tasks. Instead of adopting the common solutions to the problem by increasing the visual fidelity of synthetic images output from simulators during the training phase, we seek to tackle the problem by translating the real-world image streams back to the synthetic domain during the deployment phase, to make the robot feel at home. We propose this as a lightweight, flexible, and efficient solution for visual control, as 1) no extra transfer steps are required during the expensive training of DRL agents in simulation; 2) the trained DRL agents will not be constrained to being deployable in only one specific real-world environment; 3) the policy training and the transfer operations are decoupled, and can be conducted in parallel. Besides this, we propose a simple yet effective shift loss that is agnostic to the downstream task, to constrain the consistency between subsequent frames which is important for consistent policy outputs. We validate the shift loss for artistic style transfer for videos and domain adaptation, and validate our visual control approach in indoor and outdoor robotics experiments.
In text/plain format

Archived Files and Locations

application/pdf  10.3 MB
file_2dsyj2dy6vdvthn4own6awxkhq
arxiv.org (repository)
web.archive.org (webarchive)
Read Archived PDF
Preserved and Accessible
Type  article
Stage   submitted
Date   2018-06-25
Version   v2
Language   en ?
arXiv  1802.00265v2
Work Entity
access all versions, variants, and formats of this works (eg, pre-prints)
Catalog Record
Revision: b57e9159-a1f9-45b6-b550-11730515168b
API URL: JSON