簡易檢索 / 詳目顯示

回結果列表
研究生： 龔威勤
Kung, Wei-Ching
論文名稱： 利用分散式蒙地卡羅樹搜索合作式的資訊路徑來重建環境地圖
Environment reconstruction via cooperative information paths using distributed Monte Carlo tree search
指導教授： 蘇豐文
Soo, Von-Wun
口試委員: 朱宏國
Chu, Hung-Kuo
黃國源
Huang, Kou-Yuan
學位類別： 碩士
Master
系所名稱： 電機資訊學院 - 資訊工程學系
Computer Science
論文出版年： 2021
畢業學年度： 109
語文別： 英文
論文頁數： 22
中文關鍵詞： 資訊路徑 、合作式
外文關鍵詞： cooperative
相關次數： 點閱：2  下載：0 
分享至:
查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報

    • 資訊路徑是符合特定約束中，可將訊息搜集最大化的運動路徑，此研究目標設定在多機合作的情況下，找出能最大化未知信號場的重構結果的合作式多資訊路徑。 我們提出了一種分散式蒙地卡羅樹搜索系統，該樹通過少量仿真結果共享來減少後續步驟中的冗餘搜索。我們首先比較此方式下多機合作路徑能帶來的時間增益。與杜鵑搜索基準相比提出的方法還可以找到更好的重建方法。

    An informative path defined on a motion planning that maximizes the informative amount within the given constraint, this research target to find cooperative informative paths that can maximize the reconstructed result of the unknown signal field by combining all of them. We propose a distributed Monte-Carlo tree search system, the tree reduces the redundant search in the latter step by a small amount of the simulation result sharing. We first analyze the time gain in the way of cooperation. Then we compared it to a general cuckoo search baseline.

    摘要 Abstract Acknowledgement Contents List of Tables List of Figures 1 Introduction .............................. 1 2 Related Work .............................. 3 3 Methodology .............................. 5 3.1 Problemstatement .............................. 5 3.2 Thepredictionmodel............................. 7 3.3 dMCTSmodel ................................ 7 3.3.1 Monte-Carlotreesearch ....................... 7 3.3.2 distributedMCTS........................... 10 4 Experiments and Results 13 4.1 Experimentssetup............................... 13 4.2 Explorationtimegainthroughdistributedmethod . . . . . . . . . . . . . . 14 4.3 Environmentmapreconstructionperformance . . . . . . . . . . . . . . . . 15 5 Conclusion and Future Work 18 5.1 Conclusion .................................. 18 5.2 FutureWork.................................. 19 References.................................. 20

    [1] S. Smith B. Jones D. Rus R. Smith, M. Schwager and G. Sukhatme. ́ın. Persistent ocean monitoring with underwater gliders: Adapting sampling resolution. J.Field Robotics, 28(5):714–741, Sept 2011.

    [2] A.Gasparri E.Garone. ́ın N.Bono Rossell, R.Carpi. Information-Based Path Planning for UAV Coverage with Discrete Measurements. (Manuscript submitted for publica- tion), 2020. URL https://arxiv.org/abs/2006.11000.

    [3] J. Kim and H. I. Son. ́ın. A voronoi diagram-based workspace partition for weak cooperation of multi-robot system in orchard. IEEE Access, vol. 8:20676–20686, Jan 2020. doi: 10.1109/ACCESS.2020.2969449.

    [4] F. Hover U. Mitra G. Hollinger, B. Englot and G. Sukhatme. ́ın. Active planning for underwater inspection and the benefit of adaptivity. Int. J. Robotics Research, 32(1): 3–18, Jan. 2013. doi: 10.1177/0278364912467485.

    [5] D.GhoseR.RavichandranandK.Das. ́ın.UAVBasedSurvivorSearchduringFloods. International Conference on Unmanned Aircraft Systems (ICUAS), pages 1407–1415, 2019. doi: 10.1109/ICUAS.2019.8798127.

    [6] T.Jilek P.Sladek L.Zalud. ́ın P.Gabrlik, T.Lazna. Using anautomated heterogeneous robotic system for radiation surveys. (Manuscript submitted for publication), Jun 2020. URL https://arxiv.org/abs/2006.16066.

    [7] O. P. Kreidl A. Ghosh A. Dutta, A. Bhattacharya and P. Dasgupta ́ın. Multi- robot informative path planning in unknown environments through continuous re- gion partitioning. Int. J. Robotics Research, 17(6):1–4, Nov 2020. doi: 10.1177/ 1729881420970461.

    [8] H.Durrant-Whyte. ́ınA.Cameron.ABayesianapproachtooptimalsensorplacement. Int. J. Robotics Research, 9(5):70–88, Oct. 1990. doi: 10.1016/j.ymssp.2009.09.003.

    [9] P. T. Kabamba A. T. Klesh and ́ın A. R. Girard. Path planning for cooperative time- optimal information collection. American Control Conference, 2008:1991–1996, 2008. doi: 10.1109/ACC.2008.4586785.

    [10] S.Singh. ́ın G.Hollinger. Multi-robot coordination with periodic connectivity. Inter- national Conference on Robotics and Automation, 2010:4457–4462, 2010.

    [11] A.; Guestrin C.; Kaiser W. J.; Singh, A.; Krause and M. A. ́ın Batalin. Efficient plan- ning of informative paths for multiple robots. International joint conference on Artif- ical intelligence(IJCAI), 7:2204–2211, Jan 2007. doi: 10.5555/1625275.1625631.

    [12] A.; Guestrin, C.; Krause and A. P. ́ın Singh. Near-optimal sensor placements in gaus- sian processes. Journal of Machine Learning Research, 9:265–272, 2008.

    [13] A. R.Cassandra ́ın L.P.Kaelbling, M. L.Littman. Planning and acting in partially ob- servable stochastic domains. Artificial Intelligence, 101:99–134, May 1998. doi: 10.1016/S0004-3702(98)00023-X.

    [14] G.Sukhatme. ́ın G.Hollinger. Sampling-based robotic information gathering algo- rithms. Int. J. Robotics Research, 33(9):1271–1287, August 2014. doi: 10.1177/ 0278364914533443.

    [15] K. H. Low N. Cao and J. M. Dolan. ́ın. Multi-robot informative path planning for active sensing of environmental phenomena: A tale of two algorithms. Agents Multi- Agent Syst, pages 7–14, May 2013. doi: 10.5555/2484920.2484926.

    [16] P.Ghassemi and S.Chowdhury. Informative Path Planning with Local Penalization for Decentralized and Asynchronous Swarm Robotic Search. International Symposium on Multi-Robot and Multi-Agent Systems, pages 188–194, August. 2019. doi: 10. 1109/MRS.2019.8901084.

    [17] P. Hennig N. Lawrenc ́ın J. Gonzalez, Z. Dai. Batch bayesian optimization via local penalization. Artificial Intelligence and Statistics, page 648–657, August. 2016. doi: https://arxiv.org/abs/1505.08052.

    [18] X.Yang ́ın. Social Algorithms. Encyclopedia of Complexity and Systems Science, Apr 2017. doi: 10.1007/978-3-642-27737-5678-1.

    QR CODE