A design of optimal controllers based on a reinforcement learning method called Q-Learning is presented. Central to Q-Learning is the Q-function which is a function of the state and all input variables. This paper shows that decoupled-in-the-inputs Q-functions exist, and can be used to find the optimal controllers for each input individually. The method thus converts a multiple-variable optimization problem into much simpler single-variable optimization problems while achieving optimality. An explicit model of the system is not required to learn these decoupled Q-functions, but rather the method relies on the ability to probe the system and observe its state transition. Derived within the framework of modern control theory, the method is applicable to both linear and non-linear systems.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Input-Decoupled Q-Learning for Optimal Control


    Additional title:

    J Astronaut Sci


    Contributors:

    Published in:

    Publication date :

    2020-06-01


    Size :

    27 pages




    Type of media :

    Article (Journal)


    Type of material :

    Electronic Resource


    Language :

    English




    Input-Decoupled Q-Learning for Optimal Control

    Phan, Minh Q. / Azad, Seyed Mahdi B. | Online Contents | 2019


    Optimal fuselage aiming with decoupled short period dynamics

    JARMARK, BERNT / SPEYER, JASON | AIAA | 1988



    Multi-input/multi-output controller design for longitudinal decoupled aircraft motion

    SPEYER, J. L. / WHITE, J. E. / DOUGLAS, R. et al. | AIAA | 1984