Impact assessment, also referred to as Level 3 Data Fusion, combines information to provide a detailed understanding of how other systems and external events affect the system of interest. The information can be provided solely from the organic sensors to the system, or completely external information sources, or a combination of both. In unmanned aerial systems (UAS) the desire for autonomy is growing as the ideas for security, delivery, information gathering, and other concepts are proposed in the private sector. However, such operations require the UAS to understand its own inherent capabilities, such as its maneuverability, and to understand the world around it, such as the projected flight-paths of nearby UASs and potential obstacles. With such autonomous operations, Level 3 fusion is necessary for the UAS to interpret available information in order to make decision about its own behavior to avoid collisions and optimize is mission criteria. One technique for Level 3 fusion is evidence accrual, where low level evidence is combined through a series levels of increasingly complex steps to achieve a highly evolved understanding. For autonomous systems such as UASs, the impacts of information on a decision can vary in intensity based on a variety of issues. For example, as another UAS approaches the threat of a collision increases until one cannot react to avoid a collision or cannot modify the course to cause a collision. The modification of the weighting between the system dynamics and the impact is a time-varying system identification problem. A fuzzy-logic-based approach to evidence accrual was developed and has been applied to Level 3 fusion problems such as UAS. However, the approach has been too limited for use in autonomous systems in that it either requires operator information for adaptation of the relationships and between decisions and lower level evidence or has been static. In this effort, the development of the approach to apply to the UAS collision avoidance problem is made, by providing linear weighting adaptation for the fuzzy-logic-based evidence accrual system. The adaptation technique uses a fixed structure between the nodes that comprise the evidence and the levels of evidence accrual. The adaptation uses a Kalman filter identification technique that uses the understanding of the inherent capabilities of the UAS of interest as truth.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    Adaptation for Evidence Accrual Applied to UAS Impact Assessment


    Beteiligte:


    Erscheinungsdatum :

    01.09.2018


    Format / Umfang :

    1028024 byte





    Medientyp :

    Aufsatz (Konferenz)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch



    Adaptive UAS Route Planner Based Upon Evidence Accrual

    Kramer, Kathleen A. / Stubberud, Stephen C. | IEEE | 2019


    Forklift digital intelligent terminal with accrual function

    TAO GUANPING / WANG YANYAN / ZOU WEIRI et al. | Europäisches Patentamt | 2021

    Freier Zugriff