A Behavioral Approach to Visual Navigation with Graph Localization Networks. Chen, K., Vicente, J. D., Sepulveda, G., Xia, F., Soto, A., Vazquez, M., & Savarese, S. In RSS, 2019.
A Behavioral Approach to Visual Navigation with Graph Localization Networks [link]Paper  abstract   bibtex   3 downloads  
Inspired by research in psychology, we introduce a behavioral approach for visual navigation using topological maps. Our goal is to enable a robot to navigate from one location to another, relying only on its visual observations and the topological map of the environment. To this end, we propose using graph neural networks for localizing the agent in the map, and decompose the action space into primitive behaviors implemented as convolutional or recurrent neural networks. Using the Gibson simulator and the Stanford 2D-3D-S dataset, we verify that our approach outperforms relevant baselines and is able to navigate in both seen and unseen indoor environments.
@InProceedings{	  kevin:etal:2018,
  author	= {K. Chen and J.P De Vicente and G. Sepulveda and F. Xia and
		  A. Soto and M. Vazquez and S. Savarese},
  title		= {A Behavioral Approach to Visual Navigation with Graph
		  Localization Networks},
  booktitle	= {RSS},
  year		= {2019},
  abstract	= {Inspired by research in psychology, we introduce a
		  behavioral approach for visual navigation using topological
		  maps. Our goal is to enable a robot to navigate from one
		  location to another, relying only on its visual
		  observations and the topological map of the environment. To
		  this end, we propose using graph neural networks for
		  localizing the agent in the map, and decompose the action
		  space into primitive behaviors implemented as convolutional
		  or recurrent neural networks. Using the Gibson simulator
		  and the Stanford 2D-3D-S dataset, we verify that our
		  approach outperforms relevant baselines and is able to
		  navigate in both seen and unseen indoor environments.},
  url		= {https://graphnav.stanford.edu/}
}

Downloads: 3