Autonomous Multi-View Navigation via Deep Reinforcement Learning
Xueqin Huang, Wei Chen, Wei Zhang, Ran Song, Jiyu Cheng, Yibin Li
Abstract
In this paper, we propose a novel deep reinforcement learning (DRL) system for the autonomous navigation of mobile robots that consists of three modules: map navigation, multi-view perception and multi-branch control. Our DRL system takes as the input a routed map provided by a global planner and three RGB images captured by a multi-camera setup to gather global and local information, respectively. In particular, we present a multi-view perception module based on an attention mechanism to filter out redundant information caused by multi-camera sensing. We also replace raw RGB images with low-dimensional representations via a specifically designed network, which benefits a more robust sim2real transfer learning. Extensive experiments in both simulated and real-world scenarios demonstrate that our system outperforms state-of-the-art approaches.
BibTeX
@inproceedings{icra2021_autonomousmultiv,
title = {Autonomous Multi-View Navigation via Deep Reinforcement Learning},
author = {Xueqin Huang and Wei Chen and Wei Zhang and Ran Song and Jiyu Cheng and Yibin Li},
booktitle = {ICRA 2021},
year = {2021}
}