Previous approaches for small fixed-wing unmanned air systems that carry strapdown rather than gimbaled cameras achieved satisfactory ground target tracking performance using both standard and deep reinforcement learning algorithms. However, these approaches have significant restrictions and abstractions to the dynamics of the vehicle, such as constant airspeed and constant altitude, because the number of states and actions was necessarily limited. Thus, extensive tuning was required to obtain good tracking performance. The expansion from 4 state–action degrees of freedom to 15 enabled the agent to exploit previous reward functions that produced novel yet undesirable emergent behavior. This paper investigates the causes of and various potential solutions to undesirable emergent behavior in the ground target tracking problem. A combination of changes to the environment, reward structure, action space simplification, command rate, and controller implementation provides insight into obtaining stable tracking results. Consideration is given to reward structure selection and refinement to mitigate undesirable emergent behavior. Results presented in the paper for a simulated environment of a single unmanned air system tracking a randomly moving single ground target show that a soft actor–critic algorithm can produce feasible tracking trajectories without limiting the state space and action space, provided that the environment is properly posed.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    Design, Selection, and Evaluation of Reinforcement Learning Single Agents for Ground Target Tracking


    Beteiligte:
    Lehman, Hannah (Autor:in) / Valasek, John (Autor:in)

    Erschienen in:

    Erscheinungsdatum :

    2023-11-13


    Format / Umfang :

    18 pages




    Medientyp :

    Aufsatz (Zeitschrift)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch




    Deep Reinforcement Learning Controller for Autonomous Tracking of Evasive Ground Target

    van Wijk, David / Eves, Kameron J. / Valasek, John | AIAA | 2023




    Optimal selection of UAV for ground target tracking

    Narkiewicz, Janusz P. / Kopyt, Antoni / Radyisyewski, Pawel et al. | AIAA | 2015


    Optimal Selection of UAV for Ground Target Tracking

    Narkiewicz, Janusz / Kopyt, Antoni / Malecki, Tomasz et al. | British Library Conference Proceedings | 2015