@inproceedings{6587104359ee47c8b196b2614d174e42,
title = "Fault-Tolerant Model Predictive Control for Multirotor UAVs",
abstract = "This paper presents a method for advanced fault tolerant control (FTC) of multirotor unmanned aerial vehicles (UAVs), which includes anomaly detection on sensor measurements, fault estimation on actuators, and a robust model predictive control (MPC). To detect anomalies on the sensor measurements, an Echo State Network is used. System states and faults are estimated using an adaptive extended Kalman filter. The system is further controlled using MPC. The method is tested in numerical simulations with a hexacopter dynamic model. Simulation results show the ability of the FTC to handle failure with different even and uneven actuator faults.",
keywords = "adaptive extended Kalman filter, AI-based methods, echo state network, fault-tolerant control, hexacopter",
author = "Diget, {Emil Lykke} and Agus Hasan and Poramate Manoonpong",
note = "Funding Information: ACKNOWLEDGMENT This paper is partially funded by Equinor{\textquoteright}s gift professorship in system dynamics for development of digital twin at NTNU. Publisher Copyright: {\textcopyright} 2022 American Automatic Control Council.; 2022 American Control Conference, ACC 2022 ; Conference date: 08-06-2022 Through 10-06-2022",
year = "2022",
doi = "10.23919/ACC53348.2022.9867240",
language = "English",
series = "Proceedings of the American Control Conference",
publisher = "IEEE",
pages = "4305--4310",
booktitle = "2022 American Control Conference, ACC 2022",
address = "United States",
}