In comparison to traditional fin and rudder propulsion devices, vector thruster autonomous underwater vehicles (AUVs) are renowned for their superior maneuverability, particularly at low or zero speeds. This study introduces a control strategy for vector thruster AUVs that employs deep reinforcement learning algorithms for three-dimensional random point and trajectory tracking. Simulation tests are conducted to investigate the motion of vector propelled AUVs. The utilization of deep reinforcement learning algorithms in this research demonstrates the potential of this approach in designing control strategies for vector thruster AUVs, ultimately enhancing maneuverability and autonomy in underwater exploration.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Research on Control Problems of vectored thruster AUVs with deep reinforcement learning


    Contributors:
    Liu, Tao (author) / Huang, Junhao (author) / Zhao, Jintao (author)


    Publication date :

    2024-05-23


    Size :

    782851 byte




    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    Path guidance and attitude control of a vectored thruster AUV

    Cavallo, Emanuele / Michelini, Rinaldo C. / Filaretov, Vladimir F. et al. | Tema Archive | 2004


    Degraded Planary Tracking Control of an Omnidirectional Vectored-Thruster Aerostat

    Chen, Li / Whidborne, James F. / Dong, Qi et al. | ASCE | 2019


    Design and Modeling of a Vectored Electrospray Thruster

    Savytskyy, Ivan / Jugroot, Manish | AIAA | 2022


    ION THRUSTER FOR THRUST VECTORED PROPULSION OF A SPACECRAFT

    BULDRINI NEMBO / SEIFERT BERNHARD / REISSNER ALEXANDER | European Patent Office | 2021

    Free access

    Ion thruster for thrust vectored propulsion of a spacecraft

    BULDRINI NEMBO / SEIFERT BERNHARD / REISSNER ALEXANDER | European Patent Office | 2024

    Free access