by , , ,
Abstract:
With the expanding utilization of unmanned aerial vehicles, the importance of ensuring their safety is increasing. In this paper, we propose an adaptive model predictive control scheme to address a wide range of faults by updating both the prediction model and cost function weights according to the degree of fault. Furthermore, we discuss the potential for a learning-based approach to automate weight adaptation, aiming to reduce design cost and complexity. Simulation results validate the effectiveness of the proposed scheme, while also highlighting the necessity and potential of a weight-learning approach.
Reference:
松田 陸杜, 武石 直也, Samir Khan, 矢入 健久:予測モデルおよび重み適応型モデル予測制御によるUAVの耐故障着陸制御, In 第68回自動制御連合講演会, 名古屋, 2025.
Bibtex Entry:
@conference{MatsudaRengo2025,
title = {予測モデルおよび重み適応型モデル予測制御によるUAVの耐故障着陸制御},
author = {陸杜 松田 and 直也 武石 and Samir Khan and 健久 矢入},
labauthor = {陸杜 松田 and 直也 武石 and Samir Khan and 健久 矢入},
year = {2025},
abstract = {With the expanding utilization of unmanned aerial vehicles, the importance of ensuring their safety is increasing. In this paper, we propose an adaptive model predictive control scheme to address a wide range of faults by updating both the prediction model and cost function weights according to the degree of fault. Furthermore, we discuss the potential for a learning-based approach to automate weight adaptation, aiming to reduce design cost and complexity. Simulation results validate the effectiveness of the proposed scheme, while also highlighting the necessity and potential of a weight-learning approach.},
booktitle = {第68回自動制御連合講演会, 名古屋},
url = {https://www.jsme.or.jp/conference/rengo68/},
lang = {ja}
}