Active visual perception for mobile robot localization. Correa, J. & Soto, A. Journal of Intelligent and Robotic Systems, 58(3-4):339-354, 2010.
Paper abstract bibtex 1 download Localization is a key issue for a mobile robot, in particular in environments where a globally accurate positioning system, such as GPS, is not available. In these environments, accurate and efficient robot localization is not a trivial task, as an increase in accuracy usually leads to an impoverishment in efficiency and viceversa. Active perception appears as an appealing way to improve the localization process by increasing the richness of the information acquired from the environment. In this paper, we present an active perception strategy for a mobile robot provided with a visual sensor mounted on a pan-tilt mechanism. The visual sensor has a limited field of view, so the goal of the active perception strategy is to use the pan-tilt unit to direct the sensor to informative parts of the environment. To achieve this goal, we use a topological map of the environment and a Bayesian non-parametric estimation of robot position based on a particle filter. We slightly modify the regular implementation of this filter by including an additional step that selects the best perceptual action using Monte Carlo estimations. We understand the best perceptual action as the one that produces the greatest reduction in uncertainty about the robot position. We also consider in our optimization function a cost term that favors efficient perceptual actions. Previous works have proposed active perception strategies for robot localization, but mainly in the context of range sensors, grid representations of the environment, and parametric techniques, such as the extended Kalman filter. Accordingly, the main contributions of this work are: i) Development of a sound strategy for active selection of perceptual actions in the context of a visual sensor and a topological map; ii) Real time operation using a modified version of the particle filter and Monte Carlo based estimations; iii) Implementation and testing of these ideas using simulations and a real case scenario. Our results indicate that, in terms of accuracy of robot localization, the proposed approach decreases mean average error and standard deviation with respect to a passive perception scheme. Furthermore, in terms of efficiency, the active scheme is able to operate in real time without adding a relevant overhead to the regular robot operation.
@Article{ correa:etal:2010,
author = {J. Correa and A. Soto},
title = {Active visual perception for mobile robot localization},
journal = {Journal of Intelligent and Robotic Systems},
volume = {58},
number = {3-4},
pages = {339-354},
year = {2010},
abstract = {Localization is a key issue for a mobile robot, in
particular in environments where a globally accurate
positioning system, such as GPS, is not available. In these
environments, accurate and efficient robot localization is
not a trivial task, as an increase in accuracy usually
leads to an impoverishment in efficiency and viceversa.
Active perception appears as an appealing way to improve
the localization process by increasing the richness of the
information acquired from the environment. In this paper,
we present an active perception strategy for a mobile robot
provided with a visual sensor mounted on a pan-tilt
mechanism. The visual sensor has a limited field of view,
so the goal of the active perception strategy is to use the
pan-tilt unit to direct the sensor to informative parts of
the environment. To achieve this goal, we use a topological
map of the environment and a Bayesian non-parametric
estimation of robot position based on a particle filter. We
slightly modify the regular implementation of this filter
by including an additional step that selects the best
perceptual action using Monte Carlo estimations. We
understand the best perceptual action as the one that
produces the greatest reduction in uncertainty about the
robot position. We also consider in our optimization
function a cost term that favors efficient perceptual
actions. Previous works have proposed active perception
strategies for robot localization, but mainly in the
context of range sensors, grid representations of the
environment, and parametric techniques, such as the
extended Kalman filter. Accordingly, the main contributions
of this work are: i) Development of a sound strategy for
active selection of perceptual actions in the context of a
visual sensor and a topological map; ii) Real time
operation using a modified version of the particle filter
and Monte Carlo based estimations; iii) Implementation and
testing of these ideas using simulations and a real case
scenario. Our results indicate that, in terms of accuracy
of robot localization, the proposed approach decreases mean
average error and standard deviation with respect to a
passive perception scheme. Furthermore, in terms of
efficiency, the active scheme is able to operate in real
time without adding a relevant overhead to the regular
robot operation. },
url = {http://saturno.ing.puc.cl/media/papers_alvaro/Intell-Robots-2010.pdf}
}
Downloads: 1
{"_id":{"_str":"53427a470e946d920a0018d6"},"__v":1,"authorIDs":["32ZR23o2BFySHbtQK","3ear6KFZSRqbj6YeT","4Pq6KLaQ8jKGXHZWH","54578d9a2abc8e9f370004f0","5e126ca5a4cabfdf01000053","5e158f76f1f31adf01000118","5e16174bf67f7dde010003ad","5e1f631ae8f5ddde010000eb","5e1f7182e8f5ddde010001ff","5e26da3642065ede01000066","5e3acefaf2a00cdf010001c8","5e62c3aecb259cde010000f9","5e65830c6e5f4cf3010000e7","5e666dfc46e828de010002c9","6cMBYieMJhf6Nd58M","6w6sGsxYSK2Quk6yZ","7xDcntrrtC62vkWM5","ARw5ReidxxZii9TTZ","BjzM7QpRCG7uCF7Zf","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","vMiJzqEKCsBxBEa3v","vQE6iTPpjxpuLip2Z","wQDRsDjhgpMJDGxWX","wbNg79jvDpzX9zHLK","wk86BgRiooBjy323E","zCbPxKnQGgDHiHMWn","zf9HENjsAzdWLMDAu"],"author_short":["Correa, J.","Soto, A."],"bibbaseid":"correa-soto-activevisualperceptionformobilerobotlocalization-2010","bibdata":{"bibtype":"article","type":"article","author":[{"firstnames":["J."],"propositions":[],"lastnames":["Correa"],"suffixes":[]},{"firstnames":["A."],"propositions":[],"lastnames":["Soto"],"suffixes":[]}],"title":"Active visual perception for mobile robot localization","journal":"Journal of Intelligent and Robotic Systems","volume":"58","number":"3-4","pages":"339-354","year":"2010","abstract":"Localization is a key issue for a mobile robot, in particular in environments where a globally accurate positioning system, such as GPS, is not available. In these environments, accurate and efficient robot localization is not a trivial task, as an increase in accuracy usually leads to an impoverishment in efficiency and viceversa. Active perception appears as an appealing way to improve the localization process by increasing the richness of the information acquired from the environment. In this paper, we present an active perception strategy for a mobile robot provided with a visual sensor mounted on a pan-tilt mechanism. The visual sensor has a limited field of view, so the goal of the active perception strategy is to use the pan-tilt unit to direct the sensor to informative parts of the environment. To achieve this goal, we use a topological map of the environment and a Bayesian non-parametric estimation of robot position based on a particle filter. We slightly modify the regular implementation of this filter by including an additional step that selects the best perceptual action using Monte Carlo estimations. We understand the best perceptual action as the one that produces the greatest reduction in uncertainty about the robot position. We also consider in our optimization function a cost term that favors efficient perceptual actions. Previous works have proposed active perception strategies for robot localization, but mainly in the context of range sensors, grid representations of the environment, and parametric techniques, such as the extended Kalman filter. Accordingly, the main contributions of this work are: i) Development of a sound strategy for active selection of perceptual actions in the context of a visual sensor and a topological map; ii) Real time operation using a modified version of the particle filter and Monte Carlo based estimations; iii) Implementation and testing of these ideas using simulations and a real case scenario. Our results indicate that, in terms of accuracy of robot localization, the proposed approach decreases mean average error and standard deviation with respect to a passive perception scheme. Furthermore, in terms of efficiency, the active scheme is able to operate in real time without adding a relevant overhead to the regular robot operation. ","url":"http://saturno.ing.puc.cl/media/papers_alvaro/Intell-Robots-2010.pdf","bibtex":"@Article{\t correa:etal:2010,\n author\t= {J. Correa and A. Soto},\n title\t\t= {Active visual perception for mobile robot localization},\n journal\t= {Journal of Intelligent and Robotic Systems},\n volume\t= {58},\n number\t= {3-4},\n pages\t\t= {339-354},\n year\t\t= {2010},\n abstract\t= {Localization is a key issue for a mobile robot, in\n\t\t particular in environments where a globally accurate\n\t\t positioning system, such as GPS, is not available. In these\n\t\t environments, accurate and efficient robot localization is\n\t\t not a trivial task, as an increase in accuracy usually\n\t\t leads to an impoverishment in efficiency and viceversa.\n\t\t Active perception appears as an appealing way to improve\n\t\t the localization process by increasing the richness of the\n\t\t information acquired from the environment. In this paper,\n\t\t we present an active perception strategy for a mobile robot\n\t\t provided with a visual sensor mounted on a pan-tilt\n\t\t mechanism. The visual sensor has a limited field of view,\n\t\t so the goal of the active perception strategy is to use the\n\t\t pan-tilt unit to direct the sensor to informative parts of\n\t\t the environment. To achieve this goal, we use a topological\n\t\t map of the environment and a Bayesian non-parametric\n\t\t estimation of robot position based on a particle filter. We\n\t\t slightly modify the regular implementation of this filter\n\t\t by including an additional step that selects the best\n\t\t perceptual action using Monte Carlo estimations. We\n\t\t understand the best perceptual action as the one that\n\t\t produces the greatest reduction in uncertainty about the\n\t\t robot position. We also consider in our optimization\n\t\t function a cost term that favors efficient perceptual\n\t\t actions. Previous works have proposed active perception\n\t\t strategies for robot localization, but mainly in the\n\t\t context of range sensors, grid representations of the\n\t\t environment, and parametric techniques, such as the\n\t\t extended Kalman filter. Accordingly, the main contributions\n\t\t of this work are: i) Development of a sound strategy for\n\t\t active selection of perceptual actions in the context of a\n\t\t visual sensor and a topological map; ii) Real time\n\t\t operation using a modified version of the particle filter\n\t\t and Monte Carlo based estimations; iii) Implementation and\n\t\t testing of these ideas using simulations and a real case\n\t\t scenario. Our results indicate that, in terms of accuracy\n\t\t of robot localization, the proposed approach decreases mean\n\t\t average error and standard deviation with respect to a\n\t\t passive perception scheme. Furthermore, in terms of\n\t\t efficiency, the active scheme is able to operate in real\n\t\t time without adding a relevant overhead to the regular\n\t\t robot operation. },\n url\t\t= {http://saturno.ing.puc.cl/media/papers_alvaro/Intell-Robots-2010.pdf}\n}\n\n","author_short":["Correa, J.","Soto, A."],"key":"correa:etal:2010","id":"correa:etal:2010","bibbaseid":"correa-soto-activevisualperceptionformobilerobotlocalization-2010","role":"author","urls":{"Paper":"http://saturno.ing.puc.cl/media/papers_alvaro/Intell-Robots-2010.pdf"},"metadata":{"authorlinks":{"soto, a":"https://asoto.ing.puc.cl/publications/"}},"downloads":1},"bibtype":"article","biburl":"https://raw.githubusercontent.com/ialab-puc/ialab.ing.puc.cl/master/pubs.bib","downloads":1,"keywords":[],"search_terms":["active","visual","perception","mobile","robot","localization","correa","soto"],"title":"Active visual perception for mobile robot localization","year":2010,"dataSources":["sg6yZ29Z2xB5xP79R","sj4fjnZAPkEeYdZqL","m8qFBfFbjk9qWjcmJ","QjT2DEZoWmQYxjHXS"]}