Fixed-wing Unmanned Aerial Vehicle (UAVs) cannot hover in potential collision scenarios as flying lower than critical speeds results in the stall of the aircraft. Hovering is also not an optimal solution for Collision Avoidance (CA) as it increases mission time and is innately fuel inefficient. Thus, an algorithm that ensures the continuous motion of UAVs during CA while maintaining maximum mission performance is needed. This work proposes a decentralized Fuzzy Inference System (FIS)-based resolution algorithm that avoids potential stall speeds during CA and is optimized using a genetic algorithm to minimize path deviation during a given mission. The mathematical model of a fixed-wing UAV in this work considers coordinated turn condition. They are assumed to be equipped with range-bearing sensors to observe potential intruders in a three-dimensional environment. The UAV guidance model uses a PD control of commanded airspeed, bank angle, and flight path angle obtained for desired path variables - airspeed, heading, and altitude which in turn are obtained using lookahead point generated for given waypoints. The optimized FIS mimics the pilot behavior during collision scenarios to provide modulation parameters for desired path variables and achieves CA while ensuring minimum path deviation. FIS optimization was conducted using a pairwise conflict scenario and tested for several pairwise and multiple UAV scenarios. Multiple simulations conducted within a given functional space with random conflict scenarios of multiple aircraft yielded effective CA with minimal collisions and overall path deviation; thus, validating the effectiveness of the proposed CA algorithm.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Fuzzy Inference System-Based Collision Avoidance of Unmanned Aerial Vehicles Optimized Using Genetic Algorithm


    Additional title:

    Lect. Notes in Networks, Syst.


    Contributors:

    Conference:

    North American Fuzzy Information Processing Society Annual Conference ; 2023 ; Cincinnati, OH, USA May 31, 2023 - June 02, 2023



    Publication date :

    2023-11-25


    Size :

    12 pages





    Type of media :

    Article/Chapter (Book)


    Type of material :

    Electronic Resource


    Language :

    English