在未知環境中類神經模糊控制器於移動式機器人導航之設計

本論文共分為兩個部分，第一個部分設計了兩種模糊邏輯控制器，分別為以專家知識為基礎之模糊邏輯控制器(Expert-Knowledge-Based Fuzzy Logic Controller，EKFLC)和以障礙形態為基礎之模糊邏輯控制器(Obstacle-Configuration-Based Fuzzy Logic Controller，OCFLC)，其中EKFLC為專家知識的經驗法則所設計出的模糊規則庫；OCFLC則為設計以障礙物形態區分出多個地形，並且利用感測器和障礙物之距離和所設計出來之地形做對比度，最後與所設計之輸入角度歸屬函數做運算並推論出結論;而在此論文中這兩種模糊邏輯控制器應用於Pioneer-3DX移動式機器人，並且確認所設計之模糊邏輯控制器之導航能力及避障性能。另外提出一個新的特殊地形逃脫模式的方法，只需要利用目前障礙物與移動式機器人之間的誤差夾角和兩個門檻值來判別移動式機器人是否進入特殊地形，並且完成控制器與特殊模式之間的切換行為，解決移動式機器人陷入死巷地形的現象。我們將設計一個補償性類神經模糊控制器(Compensatory Neuro-Fuzzy Controller，CNFC)利用新型以知識為基礎之文化多策略差分進化演算法(Knowledge-Based Cultural Multi-strategy Differential Evolution，KCMDE)用於調整控制器的參數，其中我們是利用所設計的兩個模糊邏輯控制器分別產生訓練資料，提供第補償性類神經模糊控制器搭配新型進化學習演算法作為學習之用。最後我們將透過實驗結果與結論來評估兩種訓練資料之學習的效能，並比較兩種訓練資料取得方法之學習成果，並證實其效能。

關鍵字

模糊邏輯控制器；補償性類神經模糊控制器；以知識為基礎之文化多策略差分進化演算法；障礙形態；專家知識；特殊地形逃脫模式

並列摘要

This study is divided into two parts. First, we designs two fuzzy logic controllers including an expert-knowledge-based fuzzy logic controller (EKFLC) and an obstacle-configuration-based fuzzy logic controller (OCFLC). The EKFLC uses expert knowledge and experience to design fuzzy rule base for a fuzzy logic controller. The OCFLC uses “obstacle configuration” to design fuzzy rule base for pattern-mapping between quantized ultrasonic sensory data and velocity commands. The two fuzzy logic controllers are applied in mobile robots (i.e., PIONEER 3-DX) to achieve automatic navigation and obstacle avoidance capabilities. A novel escape special environment approach is proposed to let the robot can autonomously avoid some special landmarks in this project. It uses an angle between obstacle and robot, and two thresholds to determine whether that the robot enters into the special landmarks, to switch behavior-mode for solving dead-end problems. A compensatory neuro-fuzzy controller (CNFC) with a knowledge-based cultural multi-strategy differential evolution (KCMDE) is proposed to adjust system parameters. Furthermore, the two kinds of training data are produced by two fuzzy logic controllers to design the CNFC with the proposed KCMDE. Finally, the two training data are evaluated to the automatic navigation and obstacle avoidance capabilities of robots in unknown environments to achieve the objective of control of the mobile robots.

並列關鍵字

Fuzzy logic controllers ； compensatory neuro-fuzzy controller ； knowledge-based cultural multi-strategy differential evolution ； obstacle configuration ； expert knowledge ； escape special environment behavior

參考文獻

[1] J. J. Rodriguez-Andina, M. J. Moure, and M. D. Valdes, “Features, design tools, and application domains of FPGAs,” IEEE Trans. on Industrial Electronics, vol. 54, no. 4, pp. 1810-1823, Aug. 2007.

[2] E. Y. Veslin, J. Slama, M. S. Dutra, and O. Lengerke, “Motion planning on mobile robots using differential flatness,” IEEE Latin America Transactions, vol. 9, no. 7, pp. 1006-1011, Dec. 2011.

[3] A. Malinowski and H. Yu, “Comparison of embedded system design for industrial applications,” IEEE Trans. on Industrial Informatics, vol. 7, no. 2, pp. 244-254, May 2011.

[5] T. E. Mora and E. N. Sanchez, “Fuzzy logic-based real-time navigation controller for a mobile robot,” Proc. of International Conference on Intelligent Robots and Systems, Oct. 1998, pp. 612-617.

[7] G. Antonelli, S. Chiaverini, and G. Fusco, “A fuzzy-logic-based approach for mobile robot path tracking,” IEEE Trans. on Fuzzy Syst., vol. 15, no. 2, pp. 211-221, Apr. 2007.

國際替代計量

在未知環境中類神經模糊控制器於移動式機器人導航之設計

未授權

主題瀏覽