Publication Details
Hybrid Modeling Approach for Optimization Based Control of Multirotor Unmanned Aerial Vehicles
Hanák Jiří, Ing. (DCGM)
Chudý Peter, doc. Ing., Ph.D., MBA (VZ AeroWorks)
Model Predictive Control, Sparse Identification of Nonlinear Dynamics, Unmanned
Aerial Vehicle
The first principle based model synthesis is fundamental to Guidance, Navigation,
and Control (GNC) solution development and integration. Optimization techniques
such as Model Predictive Control (MPC) often rely on simplified governing
equations of the system, omitting complex interactions, which are difficult to
accurately model or pose numerical challenges for the optimization problem
solver. This paper investigates a hybrid modeling approach based on Sparse
Identification of Nonlinear Dynamics (SINDy) for local model adaptation within
the MPC framework. The presented hybrid modeling approach benefits from the known
structure of a physics-based model such that the learning process is
computationally lightweight. Numerical experiments assume a multirotor Unmanned
Aerial Vehicle (UAV) is subject to external phenomena typically encountered in
urban environments, such as ground effects or wind gusts.
@inproceedings{BUT189118,
author="Jiří {Novák} and Jiří {Hanák} and Peter {Chudý}",
title="Hybrid Modeling Approach for Optimization Based Control of Multirotor Unmanned Aerial Vehicles",
booktitle="ICAS Proceedings",
year="2024",
journal="ICAS Proceedings",
volume="9",
number="10",
pages="1--10",
publisher="International Council of the Aeronautical Sciences",
address="Florence",
issn="2958-4647",
url="https://www.icas.org/icas_archive/icas2024/data/preview/icas2024_0029.htm"
}