Stochastic Optimal Control of Unmanned Aerial Vehicles Against Cyber Attacks in the Presence of Uncertainty

2023 Fiscal Year Final Research Report

• Project/Area Number: 21K14351
• Research Category: Grant-in-Aid for Early-Career Scientists
• Allocation Type: Multi-year Fund
• Review Section: Basic Section 24010:Aerospace engineering-related
• Research Institution: Japan Aerospace EXploration Agency
• Principal Investigator: Matsuno Yoshinori 国立研究開発法人宇宙航空研究開発機構, 航空技術部門, 研究開発員 (70800845)
• Project Period (FY): 2021-04-01 – 2024-03-31
• Keywords: 確率最適制御 / 不確実性 / 飛行制御

Outline of Final Research Achievements

In this study, we develop a stochastic optimal control method with significant practical implications for effectively managing uncertainty. The primary uncertainty we address is wind error, known for its spatiotemporal correlation. We propose a practical covariance control method to handle the impact of this uncertainty on the trajectory (expected value and covariance of the trajectory). The covariance control problem, incorporating stochastic constraints, is formulated as a convex programming problem. Through extensive numerical simulations, we demonstrate the effectiveness of our proposed method. The results highlight the potential of the stochastic optimal control method to manage uncertainty and generate optimal trajectories that adhere to various state constraints.

Free Research Field

航空宇宙工学

Academic Significance and Societal Importance of the Research Achievements

本研究では，無人航空機の安全かつ効率的な運航を実現するため，コリドーと呼ばれる専用空域を設定し，不確実環境下においてあらかじめ設定されたコリドー内に留まるように飛行する最適制御手法を構築した．実際の誤差に相当する相関のある誤差を考慮した不確実環境下において，適切な最適軌道を生成できることを示した．本研究成果は，実環境における無人航空機システムの応用において，不確実性管理を強化することができ，無人航空機システムの信頼性と安全性の向上に貢献することが期待される．