YaBeSH Engineering and Technology Library

    • Journals
    • PaperQuest
    • YSE Standards
    • YaBeSH
    • Login
    View Item 
    •   YE&T Library
    • ASCE
    • Journal of Aerospace Engineering
    • View Item
    •   YE&T Library
    • ASCE
    • Journal of Aerospace Engineering
    • View Item
    • All Fields
    • Source Title
    • Year
    • Publisher
    • Title
    • Subject
    • Author
    • DOI
    • ISBN
    Advanced Search
    JavaScript is disabled for your browser. Some features of this site may not work without it.

    Archive

    Attitude Control of a Moving Mass–Actuated UAV Based on Deep Reinforcement Learning

    Source: Journal of Aerospace Engineering:;2021:;Volume ( 035 ):;issue: 002::page 04021133
    Author:
    Xiaoqi Qiu
    ,
    Changsheng Gao
    ,
    Kefan Wang
    ,
    Wuxing Jing
    DOI: 10.1061/(ASCE)AS.1943-5525.0001381
    Publisher: ASCE
    Abstract: A moving mass–actuated unmanned aerial vehicle (MAUAV) is controlled by mass sliders installed inside the airframe and has the advantages of high aerodynamic efficiency and good stealth performance. However, designing a controller for it faces severe challenges due to the strong nonlinearity and coupling of its dynamics. To this end, we proposed an attitude controller based on deep reinforcement learning for the MAUAV. It directly maps the states to the needed deflection of the actuators and is an end-to-end controller. For the sparse reward problem, the reward function required for training is reasonably designed through reward shaping to hasten the algorithm’s training speed. In training, random initialization and parameter perturbation are used to strengthen the final policy’s robustness further. The simulation results tentatively demonstrate that the proposed controller is not only robust but suboptimal. Compared with an active disturbance rejection controller (ADRC) optimized by the particle swarm algorithm, our controller still guarantees a 100% success rate in multiple unlearned scenarios, meaning it has good generalization ability.
    • Download: (2.857Mb)
    • Show Full MetaData Hide Full MetaData
    • Get RIS
    • Item Order
    • Go To Publisher
    • Statistics

      Attitude Control of a Moving Mass–Actuated UAV Based on Deep Reinforcement Learning

    URI
    https://yetl.yabesh.ir/yetl1/handle/yetl/4283335
    Collections
    • Journal of Aerospace Engineering

    Show full item record

    contributor authorXiaoqi Qiu
    contributor authorChangsheng Gao
    contributor authorKefan Wang
    contributor authorWuxing Jing
    date accessioned2022-05-07T21:06:32Z
    date available2022-05-07T21:06:32Z
    date issued2021-12-01
    identifier other(ASCE)AS.1943-5525.0001381.pdf
    identifier urihttp://yetl.yabesh.ir/yetl1/handle/yetl/4283335
    description abstractA moving mass–actuated unmanned aerial vehicle (MAUAV) is controlled by mass sliders installed inside the airframe and has the advantages of high aerodynamic efficiency and good stealth performance. However, designing a controller for it faces severe challenges due to the strong nonlinearity and coupling of its dynamics. To this end, we proposed an attitude controller based on deep reinforcement learning for the MAUAV. It directly maps the states to the needed deflection of the actuators and is an end-to-end controller. For the sparse reward problem, the reward function required for training is reasonably designed through reward shaping to hasten the algorithm’s training speed. In training, random initialization and parameter perturbation are used to strengthen the final policy’s robustness further. The simulation results tentatively demonstrate that the proposed controller is not only robust but suboptimal. Compared with an active disturbance rejection controller (ADRC) optimized by the particle swarm algorithm, our controller still guarantees a 100% success rate in multiple unlearned scenarios, meaning it has good generalization ability.
    publisherASCE
    titleAttitude Control of a Moving Mass–Actuated UAV Based on Deep Reinforcement Learning
    typeJournal Paper
    journal volume35
    journal issue2
    journal titleJournal of Aerospace Engineering
    identifier doi10.1061/(ASCE)AS.1943-5525.0001381
    journal fristpage04021133
    journal lastpage04021133-12
    page12
    treeJournal of Aerospace Engineering:;2021:;Volume ( 035 ):;issue: 002
    contenttypeFulltext
    DSpace software copyright © 2002-2015  DuraSpace
    نرم افزار کتابخانه دیجیتال "دی اسپیس" فارسی شده توسط یابش برای کتابخانه های ایرانی | تماس با یابش
    yabeshDSpacePersian
     
    DSpace software copyright © 2002-2015  DuraSpace
    نرم افزار کتابخانه دیجیتال "دی اسپیس" فارسی شده توسط یابش برای کتابخانه های ایرانی | تماس با یابش
    yabeshDSpacePersian