A Deep Learning Based Behavioral Approach to Indoor Autonomous Navigation. Sepulveda, G., Niebles, J., & Soto, A. In ICRA, 2018. Paper abstract bibtex 2 downloads We present a semantically rich graph representa- tion for indoor robotic navigation. Our graph representation encodes: semantic locations such as offices or corridors as nodes, and navigational behaviors such as enter office or cross a corridor as edges. In particular, our navigational behaviors operate directly from visual inputs to produce motor controls and are implemented with deep learning architectures. This enables the robot to avoid explicit computation of its precise location or the geometry of the environment, and enables navigation at a higher level of semantic abstraction. We evaluate the effectiveness of our representation by simulating navigation tasks in a large number of virtual environments. Our results show that using a simple sets of perceptual and navigational behaviors, the proposed approach can successfully guide the way of the robot as it completes navigational missions such as going to a specific office. Furthermore, our implementation shows to be effective to control the selection and switching of behaviors.
@InProceedings{ sepulveda:etal:2018,
author = {G. Sepulveda and JC. Niebles and A. Soto},
title = {A Deep Learning Based Behavioral Approach to Indoor
Autonomous Navigation},
booktitle = {{ICRA}},
year = {2018},
abstract = {We present a semantically rich graph representa- tion for
indoor robotic navigation. Our graph representation
encodes: semantic locations such as offices or corridors as
nodes, and navigational behaviors such as enter office or
cross a corridor as edges. In particular, our navigational
behaviors operate directly from visual inputs to produce
motor controls and are implemented with deep learning
architectures. This enables the robot to avoid explicit
computation of its precise location or the geometry of the
environment, and enables navigation at a higher level of
semantic abstraction. We evaluate the effectiveness of our
representation by simulating navigation tasks in a large
number of virtual environments. Our results show that using
a simple sets of perceptual and navigational behaviors, the
proposed approach can successfully guide the way of the
robot as it completes navigational missions such as going
to a specific office. Furthermore, our implementation shows
to be effective to control the selection and switching of
behaviors. },
url = {https://arxiv.org/pdf/1803.04119v1.pdf}
}
Downloads: 2
{"_id":"WS2YxgK43yPWTR7gZ","bibbaseid":"sepulveda-niebles-soto-adeeplearningbasedbehavioralapproachtoindoorautonomousnavigation-2018","downloads":2,"creationDate":"2018-01-20T03:05:19.925Z","title":"A Deep Learning Based Behavioral Approach to Indoor Autonomous Navigation","author_short":["Sepulveda, G.","Niebles, J.","Soto, A."],"year":2018,"bibtype":"inproceedings","biburl":"https://raw.githubusercontent.com/ialab-puc/ialab.ing.puc.cl/master/pubs.bib","bibdata":{"bibtype":"inproceedings","type":"inproceedings","author":[{"firstnames":["G."],"propositions":[],"lastnames":["Sepulveda"],"suffixes":[]},{"firstnames":["JC."],"propositions":[],"lastnames":["Niebles"],"suffixes":[]},{"firstnames":["A."],"propositions":[],"lastnames":["Soto"],"suffixes":[]}],"title":"A Deep Learning Based Behavioral Approach to Indoor Autonomous Navigation","booktitle":"ICRA","year":"2018","abstract":"We present a semantically rich graph representa- tion for indoor robotic navigation. Our graph representation encodes: semantic locations such as offices or corridors as nodes, and navigational behaviors such as enter office or cross a corridor as edges. In particular, our navigational behaviors operate directly from visual inputs to produce motor controls and are implemented with deep learning architectures. This enables the robot to avoid explicit computation of its precise location or the geometry of the environment, and enables navigation at a higher level of semantic abstraction. We evaluate the effectiveness of our representation by simulating navigation tasks in a large number of virtual environments. Our results show that using a simple sets of perceptual and navigational behaviors, the proposed approach can successfully guide the way of the robot as it completes navigational missions such as going to a specific office. Furthermore, our implementation shows to be effective to control the selection and switching of behaviors. ","url":"https://arxiv.org/pdf/1803.04119v1.pdf","bibtex":"@InProceedings{\t sepulveda:etal:2018,\n author\t= {G. Sepulveda and JC. Niebles and A. Soto},\n title\t\t= {A Deep Learning Based Behavioral Approach to Indoor\n\t\t Autonomous Navigation},\n booktitle\t= {{ICRA}},\n year\t\t= {2018},\n abstract\t= {We present a semantically rich graph representa- tion for\n\t\t indoor robotic navigation. Our graph representation\n\t\t encodes: semantic locations such as offices or corridors as\n\t\t nodes, and navigational behaviors such as enter office or\n\t\t cross a corridor as edges. In particular, our navigational\n\t\t behaviors operate directly from visual inputs to produce\n\t\t motor controls and are implemented with deep learning\n\t\t architectures. This enables the robot to avoid explicit\n\t\t computation of its precise location or the geometry of the\n\t\t environment, and enables navigation at a higher level of\n\t\t semantic abstraction. We evaluate the effectiveness of our\n\t\t representation by simulating navigation tasks in a large\n\t\t number of virtual environments. Our results show that using\n\t\t a simple sets of perceptual and navigational behaviors, the\n\t\t proposed approach can successfully guide the way of the\n\t\t robot as it completes navigational missions such as going\n\t\t to a specific office. Furthermore, our implementation shows\n\t\t to be effective to control the selection and switching of\n\t\t behaviors. },\n url\t\t= {https://arxiv.org/pdf/1803.04119v1.pdf}\n}\n\n","author_short":["Sepulveda, G.","Niebles, J.","Soto, A."],"key":"sepulveda:etal:2018","id":"sepulveda:etal:2018","bibbaseid":"sepulveda-niebles-soto-adeeplearningbasedbehavioralapproachtoindoorautonomousnavigation-2018","role":"author","urls":{"Paper":"https://arxiv.org/pdf/1803.04119v1.pdf"},"metadata":{"authorlinks":{"soto, a":"https://asoto.ing.puc.cl/publications/"}},"downloads":2},"search_terms":["deep","learning","based","behavioral","approach","indoor","autonomous","navigation","sepulveda","niebles","soto"],"keywords":[],"authorIDs":["32ZR23o2BFySHbtQK","3ear6KFZSRqbj6YeT","4Pq6KLaQ8jKGXHZWH","54578d9a2abc8e9f370004f0","5e126ca5a4cabfdf01000053","5e158f76f1f31adf01000118","5e16174bf67f7dde010003ad","5e1f631ae8f5ddde010000eb","5e1f7182e8f5ddde010001ff","5e26da3642065ede01000066","5e3acefaf2a00cdf010001c8","5e62c3aecb259cde010000f9","5e65830c6e5f4cf3010000e7","5e666dfc46e828de010002c9","6cMBYieMJhf6Nd58M","6w6sGsxYSK2Quk6yZ","7xDcntrrtC62vkWM5","ARw5ReidxxZii9TTZ","DQ4JRTTWkvKXtCNCp","GbYBJvxugXMriQwbi","HhRoRmBvwWfD4oLyK","JFk6x26H6LZMoht2n","JvArGGu5qM6EvSCvB","LpqQBhFH3PxepH9KY","MT4TkSGzAp69M3dGt","QFECgvB5v2i4j2Qzs","RKv56Kes3h6FwEa55","Rb9TkQ3KkhGAaNyXq","RdND8NxcJDsyZdkcK","SpKJ5YujbHKZnHc4v","TSRdcx4bbYKqcGbDg","W8ogS2GJa6sQKy26c","WTi3X2fT8dzBN5d8b","WfZbctNQYDBaiYW6n","XZny8xuqwfoxzhBCB","Xk2Q5qedS5MFHvjEW","bbARiTJLYS79ZMFbk","cBxsyeZ37EucQeBYK","cFyFQps7W3Sa2Wope","dGRBfr8zhMmbwK6eP","eRLgwkrEk7T7Lmzmf","fMYSCX8RMZap548vv","g6iKCQCFnJgKYYHaP","h2hTcQYuf2PB3oF8t","h83jBvZYJPJGutQrs","jAtuJBcGhng4Lq2Nd","pMoo2gotJcdDPwfrw","q5Zunk5Y2ruhw5vyq","rzNGhqxkbt2MvGY29","uC8ATA8AfngWpYLBq","uoJ7BKv28Q6TtPmPp","vMiJzqEKCsBxBEa3v","vQE6iTPpjxpuLip2Z","wQDRsDjhgpMJDGxWX","wbNg79jvDpzX9zHLK","wk86BgRiooBjy323E","zCbPxKnQGgDHiHMWn","zf9HENjsAzdWLMDAu"],"dataSources":["3YPRCmmijLqF4qHXd","sg6yZ29Z2xB5xP79R","sj4fjnZAPkEeYdZqL","m8qFBfFbjk9qWjcmJ","QjT2DEZoWmQYxjHXS"]}