臺灣博碩士論文加值系統

本文主要探討類神經網路控制器於飛機自動著陸系統 (Automatic Landing System) 之應用。目前國內外有關飛行控制的研究，其控制器的設計大多是以傳統式現代控制理論為基礎，輔以最佳化控制理論或適應控制理論而成。因為系統的簡化及線性化，使得飛機的飛行控制被劃分成許多不同的飛行區間，每一區間都有其預設的飛行條件，所以使得控制器之設計與製造變得極為複雜且不經濟。再者，飛機空難的發生，大部份是因人為的疏失及天候的影響。在降落時，當實際飛行狀況超出原先自動著陸器設計的範圍時，駕駛員就須接手操控，但大多數的駕駛員均沒有碰過干擾極大的剪風與亂流的經驗。是故，設計一智慧型的自動著陸控制器，以取代目前傳統式控制方法的自動著陸系統，是有其必要性及迫切性。本文用一多層前饋類神經網路來學習飛機於風擾中之控制能力，學習法則為倒傳遞演算法，其學習模式採循序模式。模擬結果顯示，此類神經網路控制器能於特定的亂流 (wind turbulence) 與剪風 (wind shear) 中，導引飛機自動著陸，且符合安全降落之定義範圍內。於模擬的過程中，發現隱藏層所需的節點數量以及應採用的訓練方式，所必須達到的條件，否則將難以模擬成功。此外，模擬經驗亦指出會使安全著陸變得很困難的風擾風速，足以供飛機駕駛員判定應否為維護飛航安全而中止著陸的參考。

The purpose of this thesis is to apply neural network controller to aircraft automatic landing system. Most of the studies regarding flight control are based on conventional modern control theories with optimal or adaptive control methods. Typically these techniques involve linearizing the aircraft dynamics about several operating conditions through out the flight envelope, designing linear controllers for each condition, and gain-scheduling approach. The disadvantages are that it require extensive computations, memory utilization, and memory space. Furthermore, conventional controllers are not capable of handling environmental disturbances (such as wind shear and wind turbulence). It is noted that existing automatic landing system work reliably only within a carefully specified operational safety envelope. A conventional autopilot has difficulties in severe wind disturbances during the landing phase. Thus, to increase the safety of landing it would be desirable if new method could expand the operational envelope to include safe responses under different environmental conditions. Artificial neural networks, which have the ability to approximate continuous nonlinear functions, offer the potential to overcome these problems. In this study, a multi layer feedforward neural network is applied to the aircraft landing control. The neural network controller can learn different flight conditions off-line. Simulation results show that the trained controller can guide the aircraft landing safely through wind disturbances. The analysis of using different numbers of hidden unit are included. Through the simulations, the combination of the training data and scaling skill are found. Also, strength ranges of wind turbulence and wind shear for a safety landing are suggested.

摘要 (中文) i
摘要 (英文) ii
誌謝 iii
目錄 iv
圖目錄 vii
表目錄 xiii
第一章 導論 1
1.1 研究動機 1
1.2 文獻回顧 2
1.3 論文貢獻 3
1.4 論文大綱 4
第二章 類神經網路理論與應用 5
2.1 類神經網路系統的優點 5
2.2 類神經元模型 6
2.3 類神經網路架構 9
2.4 倒傳遞演算法 12
2.5 倒傳遞演算法的理論推導 13
第三章 飛行控制 17
3.1 飛機著陸分析 17
3.2 飛機的動態模式 18
3.3 安全降落的定義 22
3.4 風擾的數學模式 23
3.4.1 亂流 23
3.4.2 剪風 25
第四章 類神經網路控制與模擬 28
4.1 控制方式 28
4.2 模擬結果 30
4.2.1 傳統控制器 30
4.2.2 預訓練結果 32
4.2.3 類神經網路控制器模擬結果 35
4.3 結果討論 37
第五章 亂流中著陸控制 38
5.1 控制方式 38
5.2 傳統控制器於亂流之模擬 39
5.3 類神經網路控制器於亂流之模擬 49
5.3.1 不同節點數所造成的差異 49
5.3.2 不同風速之亂流的預訓練 59
5.3.3 不同亂流之著陸控制 62
5.4 結果討論 63
第六章 剪風中著陸控制 65
6.1 傳統控制器於剪風之模擬 65
6.2 類神經網路控制器於剪風之模擬 73
6.2.1 剪風下的預訓練 73
6.2.2 不同剪風之類神經網路著陸控制 80
6.2.3 單一類神經網路控制器於不同剪風之著陸模擬 80
6.3 結果討論 81
第七章 結論與建議 82
參考文獻 84

1. Federal Aviation Administration, “Automatic Landing Systems,” AC 20-57A, Jan. 1971.
2. H. Buschek and A.J. Calise, “Uncertainty Modeling and Fixed-Order Controller Design for a Hypersonic Vehicle Model,” Journal of Guidance, Control, and Dynamics, vol. 20, no. 1, pp. 42-48, 1997.
3. A.J. Calise and R.T. Rysdyk, “Nonlinear Adaptive Flight Control Using Neural Networks,” IEEE Control Systems Magazine, vol. 18, no. 6, pp. 14-25, 1998.
4. P.J. Werbos, “Neural Networks and Flight Control: Overview of Capabilities and Emerging Applications,” Proc. Of the AIAA Guidance, Navigation, and Control Conf., pp. 912-919, 1995.
5. M.R. Napolitano and M. Kincheloe, “On-Line Learning Neural-Network Controller for Autopilot Systems,” Journal of Guidance, Control, and Dynamics, vol. 18, no. 5, pp. 1008-1015, 1995.
6. 陳柏熹, “The Application of Neural Network on Longitudinal Flight Control System Design,” M.S. thesis, 中興大學, 1994.
7. 謝成德, “Gain Scheduling of Flight Control System with Fuzzy Logic Model,” M.S. thesis, 交通大學, 1996.
8. 譚俊豪, “模糊飛行控制系統的發展及應用於無人飛機之研究,” 國科會研究報告, NSC82-042-E006-148, 1993.
9. M.B. McFarland and A.J. Calise, “Robust Adaptive Control of Uncertain Nonlinear Systems Using Neural Networks,” Proc. American Control Conf., pp. 1996-2000, 1997.
10. Special Issue on Neural Network Hardware, IEEE Trans. Neural Networks, vol. 3, no. 3, 1992.
11. C.E. Cohen, et al., “Automatic Landing of a 737 Using GNSS Integrity Beacons,” Proc. ISPA, 1995.
12. DDC-I, “Advanced Auto Landing System from Swiss Federal Aircraft Factory,” Real-Time Journal, Sprint, 1995.
13. S. Asai, et al., “Development of Flight Control System for Automatic Landing Flight Experiment,” Mitsubishi Heavy Industries Technical Review, vol. 34, no. 3, 1997.
14. D.N. Kaufmann and B.D. McNally, “Flight Test Evaluation of the Stanford University and United Airlines Differential GPS Category III Automatic Landing System,” NASA Technical Memorandum 110354, June 1995.
15. M.G. Cooper, “Genetic Design of Rule-Based Fuzzy Controllers,” Ph.D. dissertation, University of California, Los Angeles, 1995.
16. C.C. Jorgensen and C. Schley, “A Neural Network Baseline Problem for Control of Aircraft Flare and Touchdown,” Neural Networks for Control, pp. 403-425, 1991.
17. K. Hornik, M. Stinchombe and H. White, “Multilayer Feedforward Networks are Universal Approximators,” Neural Networks, vol. 2, no. 5, pp.359-366, 1989.
18. R.M. Sanner and J.E. Slotine, “Gaussian Networks for Direct Adaptive Control,” IEEE Trans. Neural Networks, vol. 3, no. 6, pp. 837-863, 1992.
19. M.S. Lan, “Adaptive Control of Unknown Dynamical System via Neural Network Approach,” Proc. American Control Conf., pp. 910-915, 1991.
20. Y.H. Chen and S. Pandey, “Robust Control Strategy for Take-Off Performance in a Windshear,” Optimal Control Applications and Method, vol. 10, no. 1, pp. 65-79, 1989.
21. S. Haykin, Neural Networks: A Comprehensive Foundation, Printice-Hall, Inc., 1999.
22. Kuschewski J. G. and Zak S. H. “Application of Feedforward Neural Network to Dynamical System Identification and Control” IEEE Transactions on Control System Technology, Vol. 1. No 1, pp. 37-49, March 1993.
23. W.T. Miller, R.S. Sutton, and P.J. Werbos, “Neural Networks for Control”, the MIT Press, 1991.
24. 蘇木春, 張孝德, “機械學習：類神經網路、模糊系統以及基因演算法”, 全華科技圖書股份有限公司, 1999.
25. R.C. Nelson, “Flight Stability and Automatic Control”, 4th edition, pp.314-315, Mcgraw-Hill International Editions, 1998.
26. R.Bulirsch, F.Montrone, and H. Pesch, “Abort Landing of the Presence of Windshear as a Minimax Optimal Control Problem,” vol. 1, PP.1-23, 1991.