Hybrid Modeling Approach for Optimization Based Control of Multirotor Unmanned Aerial Vehicles

Hybrid Modeling Approach for Optimization Based Control of Multirotor Unmanned Aerial Vehicles

Paper in proceedings (conference paper)

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.

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.

Model Predictive Control, Sparse Identification of Nonlinear Dynamics, Unmanned Aerial Vehicle

Model Predictive Control, Sparse Identification of Nonlinear Dynamics, Unmanned Aerial Vehicle

NOVÁK, J.; HANÁK, J.; CHUDÝ, P.

2025

08.10.2024

International Council of the Aeronautical Sciences

Florence

ICAS Proceedings

2958-4647

ICAS Proceedings

9

10

Federal Republic of Germany

1

10

10

https://www.icas.org/icas_archive/icas2024/data/preview/icas2024_0029.htm