Deep Reinforcement Learning Robot for Search and Rescue Applications: Exploration in Unknown Cluttered Environments. Niroui, F., Zhang, K., Kashino, Z., & Nejat, G. IEEE Robotics and Automation Letters, 4(2):610–617, April, 2019. doi abstract bibtex 2 downloads Rescue robots can be used in urban search and rescue (USAR) applications to perform the important task of exploring unknown cluttered environments. Due to the unpredictable nature of these environments, deep learning techniques can be used to perform these tasks. In this letter, we present the first use of deep learning to address the robot exploration task in USAR applications. In particular, we uniquely combine the traditional approach of frontier-based exploration with deep reinforcement learning to allow a robot to autonomously explore unknown cluttered environments. Experiments conducted with a mobile robot in unknown cluttered environments of varying sizes and layouts showed that the proposed exploration approach can effectively determine appropriate frontier locations to navigate to, while being robust to different environment layouts and sizes. Furthermore, a comparison study with other frontier exploration approaches showed that our learning-based frontier exploration technique was able to explore more of an environment earlier on, allowing for potential identification of a larger number of victims at the beginning of the time-critical exploration task.
@article{niroui_deep_2019,
title = {Deep {Reinforcement} {Learning} {Robot} for {Search} and {Rescue} {Applications}: {Exploration} in {Unknown} {Cluttered} {Environments}},
volume = {4},
issn = {2377-3766},
shorttitle = {Deep {Reinforcement} {Learning} {Robot} for {Search} and {Rescue} {Applications}},
doi = {10.1109/LRA.2019.2891991},
abstract = {Rescue robots can be used in urban search and rescue (USAR) applications to perform the important task of exploring unknown cluttered environments. Due to the unpredictable nature of these environments, deep learning techniques can be used to perform these tasks. In this letter, we present the first use of deep learning to address the robot exploration task in USAR applications. In particular, we uniquely combine the traditional approach of frontier-based exploration with deep reinforcement learning to allow a robot to autonomously explore unknown cluttered environments. Experiments conducted with a mobile robot in unknown cluttered environments of varying sizes and layouts showed that the proposed exploration approach can effectively determine appropriate frontier locations to navigate to, while being robust to different environment layouts and sizes. Furthermore, a comparison study with other frontier exploration approaches showed that our learning-based frontier exploration technique was able to explore more of an environment earlier on, allowing for potential identification of a larger number of victims at the beginning of the time-critical exploration task.},
number = {2},
journal = {IEEE Robotics and Automation Letters},
author = {Niroui, F. and Zhang, K. and Kashino, Z. and Nejat, G.},
month = apr,
year = {2019},
keywords = {Autonomous agents, Computer architecture, Layout, Microprocessors, Navigation, Robot sensing systems, Task analysis, USAR applications, deep learning in robotics and automation, deep reinforcement learning robot, learning (artificial intelligence), learning-based frontier exploration technique, mobile robot, path planning, rescue robots, robot exploration task, search and rescue robots, time-critical exploration task, unknown cluttered environments, urban search and rescue applications},
pages = {610--617},
}
Downloads: 2
{"_id":"jq8m87pFuvuN7YNbC","bibbaseid":"niroui-zhang-kashino-nejat-deepreinforcementlearningrobotforsearchandrescueapplicationsexplorationinunknownclutteredenvironments-2019","authorIDs":["5cd1ce0e9c890bee0100011a","TiQ2byWdRvZBJ87S6","ei4phBGEgFBxeQyai"],"author_short":["Niroui, F.","Zhang, K.","Kashino, Z.","Nejat, G."],"bibdata":{"bibtype":"article","type":"article","title":"Deep Reinforcement Learning Robot for Search and Rescue Applications: Exploration in Unknown Cluttered Environments","volume":"4","issn":"2377-3766","shorttitle":"Deep Reinforcement Learning Robot for Search and Rescue Applications","doi":"10.1109/LRA.2019.2891991","abstract":"Rescue robots can be used in urban search and rescue (USAR) applications to perform the important task of exploring unknown cluttered environments. Due to the unpredictable nature of these environments, deep learning techniques can be used to perform these tasks. In this letter, we present the first use of deep learning to address the robot exploration task in USAR applications. In particular, we uniquely combine the traditional approach of frontier-based exploration with deep reinforcement learning to allow a robot to autonomously explore unknown cluttered environments. Experiments conducted with a mobile robot in unknown cluttered environments of varying sizes and layouts showed that the proposed exploration approach can effectively determine appropriate frontier locations to navigate to, while being robust to different environment layouts and sizes. Furthermore, a comparison study with other frontier exploration approaches showed that our learning-based frontier exploration technique was able to explore more of an environment earlier on, allowing for potential identification of a larger number of victims at the beginning of the time-critical exploration task.","number":"2","journal":"IEEE Robotics and Automation Letters","author":[{"propositions":[],"lastnames":["Niroui"],"firstnames":["F."],"suffixes":[]},{"propositions":[],"lastnames":["Zhang"],"firstnames":["K."],"suffixes":[]},{"propositions":[],"lastnames":["Kashino"],"firstnames":["Z."],"suffixes":[]},{"propositions":[],"lastnames":["Nejat"],"firstnames":["G."],"suffixes":[]}],"month":"April","year":"2019","keywords":"Autonomous agents, Computer architecture, Layout, Microprocessors, Navigation, Robot sensing systems, Task analysis, USAR applications, deep learning in robotics and automation, deep reinforcement learning robot, learning (artificial intelligence), learning-based frontier exploration technique, mobile robot, path planning, rescue robots, robot exploration task, search and rescue robots, time-critical exploration task, unknown cluttered environments, urban search and rescue applications","pages":"610–617","bibtex":"@article{niroui_deep_2019,\n\ttitle = {Deep {Reinforcement} {Learning} {Robot} for {Search} and {Rescue} {Applications}: {Exploration} in {Unknown} {Cluttered} {Environments}},\n\tvolume = {4},\n\tissn = {2377-3766},\n\tshorttitle = {Deep {Reinforcement} {Learning} {Robot} for {Search} and {Rescue} {Applications}},\n\tdoi = {10.1109/LRA.2019.2891991},\n\tabstract = {Rescue robots can be used in urban search and rescue (USAR) applications to perform the important task of exploring unknown cluttered environments. Due to the unpredictable nature of these environments, deep learning techniques can be used to perform these tasks. In this letter, we present the first use of deep learning to address the robot exploration task in USAR applications. In particular, we uniquely combine the traditional approach of frontier-based exploration with deep reinforcement learning to allow a robot to autonomously explore unknown cluttered environments. Experiments conducted with a mobile robot in unknown cluttered environments of varying sizes and layouts showed that the proposed exploration approach can effectively determine appropriate frontier locations to navigate to, while being robust to different environment layouts and sizes. Furthermore, a comparison study with other frontier exploration approaches showed that our learning-based frontier exploration technique was able to explore more of an environment earlier on, allowing for potential identification of a larger number of victims at the beginning of the time-critical exploration task.},\n\tnumber = {2},\n\tjournal = {IEEE Robotics and Automation Letters},\n\tauthor = {Niroui, F. and Zhang, K. and Kashino, Z. and Nejat, G.},\n\tmonth = apr,\n\tyear = {2019},\n\tkeywords = {Autonomous agents, Computer architecture, Layout, Microprocessors, Navigation, Robot sensing systems, Task analysis, USAR applications, deep learning in robotics and automation, deep reinforcement learning robot, learning (artificial intelligence), learning-based frontier exploration technique, mobile robot, path planning, rescue robots, robot exploration task, search and rescue robots, time-critical exploration task, unknown cluttered environments, urban search and rescue applications},\n\tpages = {610--617},\n}\n\n","author_short":["Niroui, F.","Zhang, K.","Kashino, Z.","Nejat, G."],"key":"niroui_deep_2019","id":"niroui_deep_2019","bibbaseid":"niroui-zhang-kashino-nejat-deepreinforcementlearningrobotforsearchandrescueapplicationsexplorationinunknownclutteredenvironments-2019","role":"author","urls":{},"keyword":["Autonomous agents","Computer architecture","Layout","Microprocessors","Navigation","Robot sensing systems","Task analysis","USAR applications","deep learning in robotics and automation","deep reinforcement learning robot","learning (artificial intelligence)","learning-based frontier exploration technique","mobile robot","path planning","rescue robots","robot exploration task","search and rescue robots","time-critical exploration task","unknown cluttered environments","urban search and rescue applications"],"metadata":{"authorlinks":{"nejat, g":"https://bibbase.org/show?bib=https://dblp.org/pid/14/3835.bib","kashino, z":"https://kashino.info/jp/publications/"}},"downloads":2},"bibtype":"article","biburl":"https://api.zotero.org/users/2805047/collections/ICBS3JG5/items?key=vu5JAgEl9brxvxihKOTCpenn&format=bibtex&limit=100","creationDate":"2019-05-07T18:27:26.715Z","downloads":2,"keywords":["autonomous agents","computer architecture","layout","microprocessors","navigation","robot sensing systems","task analysis","usar applications","deep learning in robotics and automation","deep reinforcement learning robot","learning (artificial intelligence)","learning-based frontier exploration technique","mobile robot","path planning","rescue robots","robot exploration task","search and rescue robots","time-critical exploration task","unknown cluttered environments","urban search and rescue applications"],"search_terms":["deep","reinforcement","learning","robot","search","rescue","applications","exploration","unknown","cluttered","environments","niroui","zhang","kashino","nejat"],"title":"Deep Reinforcement Learning Robot for Search and Rescue Applications: Exploration in Unknown Cluttered Environments","year":2019,"dataSources":["ypSzwLqyQJEZpq9yK","Kt7fjGm7dcJM3Q3uK"]}