Result Details
Weight-varying Model Predictive Control for Coupled Cyber-Physical Systems: Aerial Grasping Study
Chudý Peter, doc. Ing., Ph.D., MBA
Hanák Jiří, Ing., Ph.D., DCGM (FIT)
Advances in numerical optimization methods have enabled utilization of Nonlinear Model Predictive Control (NMPC) for increasingly complex cyber-physical systems. Tasks requiring interaction of physically connected or unconnected multi-degree of freedom dynamical systems are termed as coupled since they require mutual coordination to achieve a desired goal. This paper is focused on studying an aerial grasping problem composed of an Unmanned Aerial Vehicle (UAV) and a robotic manipulator designed for tasks such as transportation or infrastructure repair. A weight-varying approach based on Reinforecement Learning (RL) is proposed to learn a policy adapting the objective parametrization online based on a set of specified features. Considering that computing the control input actions is streamlined to a proven nonlinear optimization solver, the learning process takes less computational resources compared to learning a policy directly at the control input level.
Model Predictive Control; Unmanned Aerial Vehicle; Reinforcement Learning
@inproceedings{BUT189120,
author="Jiří {Novák} and Peter {Chudý} and Jiří {Hanák}",
title="Weight-varying Model Predictive Control for Coupled Cyber-Physical Systems: Aerial Grasping Study",
booktitle="Machine Learning, Optimization, and Data Science",
year="2025",
journal="Lecture Notes in Computer Science",
pages="13--27",
publisher="Springer Nature Switzerland AG",
address="Castiglione della Pescaia",
doi="10.1007/978-3-031-82481-4\{_}2",
isbn="978-3-031-82481-4",
issn="0302-9743",
url="https://link.springer.com/chapter/10.1007/978-3-031-82481-4_2"
}