臺灣博碩士論文加值系統

本研究結合了細胞集合(Cell Assemblies)與模糊邏輯理論，並依照人類的感知與行為模式，選用多種感測設備，而設計出一套智慧型移動系統。為了能安全並有效率的於險惡的災區中移動，我們依據需求的不同，設計了兩個模型，並融合在一起使用。具體的來說，基於模糊邏輯設計的「機器人移動」模型是用於狹窄或黑暗的環境之中，而在空曠的場合，則可透過「智慧型認知」模型及視訊攝影機的協助，讓機器人能有智慧的改變方向。
此系統中最特別的是「智慧型認知」模型所使用的細胞集合理論。此理論是以具有fLIF特性的神經元細胞為基礎。其不僅融入了工作記憶體與長期記憶體的概念，以仿效人類認知的過程，亦能透過近似人類學習的方式，不斷擴充並延展系統的功能。
本研究透過巧妙的結合不同的感測裝置，可大量降低使用視訊及影像處理的次數，且能避開一些常因誤判，產生的潛在危險。此系統經過一些構想的模擬環境測試，實驗結果證明，此系統不僅能產生出正確的移動指令，並能有效的縮短移動路徑，以滿足救災的需求。

In this study, an intelligent navigation system is developed by using fuzzy logic and Cell Assemblies (CAs) approaches and various kinds of sensors in accordance with human behavior are imported. Since disaster areas may be extremely dangerous and broad, it is important to make the robot’s movement be safer and more efficient; therefore, two models based on different purposes are designed. More specifically, ‘Robot navigation’ model implemented by fuzzy logic is applied to a narrow or dark place while ‘intelligent cognitive’ model is utilized in intelligent direction change in an open area by integrating vision camera.
In particular, the intelligent cognitive model that is implemented by CAs with fatiguing Leaky Integrate and Fire (fLIF) neurons absorb the ideals of working and long-term memory to imitate the cognitive processes of human. Additionally, there is no difficulty to combine and expand this model to be multi-functional, which makes it the same as human learning.
In this study, the clever combination of sensors significantly reduces the frequency of using vision and imaging processing. Furthermore, the combined system can avoid potential risk and efficiently shorten the motion time by importing the video camera. This system has been tested in several simulated schemes of environments and the experimental results have proved that it can not only produce right action commands regarding different schemes but also improve the motion path effectively to meet the requirements of disaster relieving.

目次

圖次 VIII
表次 IX

第一章 緒論 1
第二章 文獻探討 4
2.1. 環境感知相關理論 4
2.2. 機器人移動與人工智慧理論 6
2.2.1. 細胞集合網路 8
第三章 系統架構 15
3.1. 系統組態 15
3.2. 系統設計 16
3.3. 機器人移動模型 18
3.3.1. 模型架構 19
3.3.2. 模糊化 21
3.3.3. 推論規則庫 24
3.3.4. 解模糊化 26
3.4. 智慧型認知模型 28
3.4.1. 影像處理 29
3.4.2. 模型架構 32
3.4.3. 連線與參數設定 35
第四章 系統整合與實作 39
4.1. 預先處理工作 39
4.2. 導入智慧型認知模型 41
4.3. 智慧型認知模型工作機制 43
4.4. 實例與特性 45
第五章 實驗結果 49
5.1. 第一階段 51
5.2. 第二階段 52
5.3. 實際測試與小結 57
第六章 結論及未來展望 60
參考文獻 62

圖次
圖 一、設計流程圖 17
圖 二、模糊推論結構 21
圖 三、歸屬函數 22
圖 四、影像處理流程圖 30
圖 五、3D虛擬環境示意圖 30
圖 六、經特徵掃描後之場景 32
圖 七、經轉換成二值圖之場景 32
圖 八、智慧型認知模型之架構圖 34
圖 九、目標方位與處理優先順序 40
圖 十、系統整合工作流程圖 43
圖 十一、3D虛擬場景俯視圖 51
圖 十二、機器人移動路徑圖 52
圖 十三、於虛擬環境截取及經處理後之影像 53
圖 十四、「智慧型認知」模型執行過程 56
圖 十五、雷射掃描器偵測之障礙物其位置描繪圖 58

表次
表 一、推論規則庫 24
表 二、‘輸入’及‘因子’模組間連線的設定值 37
表 三、‘因子’及‘行為指令’模組連線的相關設定 38

Aguirre,E., & Gonzalez, A. (2003). A fuzzy perceptual model for ultrasound sensors applied to intelligent navigation of mobile robots. Applied Intelligence, 19, 171-187.
Alnajjar, F., & Murase, K. (2006). Self-organization of spiking neural network that generates autonomous behavior in a real mobile robot. International Journal of Neural Systems, 16, 229-239.
Ashley, T., Jonathan. R., & Kane. U. (2005). Is the Sun Too Bright in Queensland? An Approach to Robust Outdoor Colour Beacon Detection. In Proceedings of the Australasian Confernece on Robotics and Automation. Sydney, Australia.
Bartlett, F. C. (1932). Remembering: A Study in Experimental and Social Psychology. New York: Cambridge University Press.
Brogardh, T. (2007). Present and future robot control development - An industrial perspective. Annual Reviews in Control, 31, 69-79.
CRASAR (2004). History of rescue robots. http://crasar.cse.tamu.edu/rescuerobots/history.htm
Fan, Y. & Huyck, C. R. (2008). Implementation of Finite State Automata using fLIF Neurons. In IEEE SMC UK&RI 6th Conference on Cybernetic Systems London, UK.
Hebb, D. O. (2002). The Organization of Behavior. (New edition ed.) Lawrence Erlbaum.
Hoffmann, H. (2007). Perception through visuomotor anticipation in a mobile robot. Neural Networks, 20, 22-33.
Huyck, C. R., & Belavkin, R. (2006). Counting with Neurons: Rule Application with Nets of Fatiguing Leaky Integrate and Fire Neurons. In The 7th International Conference on Cognitive Modelling. Trieste, Italy.
Huyck, C. R., Diaper, D., Belavkin, R., & Kenny, I. (2006). Vision in an Agent based on Fatiguing Leaky Integrate and Fire Neuron. In IEEE Systems, Man and Cybernetics Society 5th Chapter Conference on Advances in Cybernetic Systems. Sheffield.
Huyck, C. R., & Fan, Y. (2007). Parsing with fLIF Neurons. In IEEE SMC UK&RI 6th Conference on Cybernetic Systems Cybernetics Society. UK.
Huyck, C. R. (2007). Creating hierarchical categories using cell assemblies. Connection Science, 19, 1-24.
Huyck, C. R. & Orengo, V. (2005). Information retrieval and categorisation using a cell assembly network. Neural Computing & Applications, 14, 282-289.
Kurby, C. A., & Zacks, J. M. (2008). Segmentation in the perception and memory of events. Trends in Cognitive Sciences, 12, 72-79.
Lemaire, T., Berger, C., Jung, I. K., & Lacroix, S. (2007). Vision-based SLAM: Stereo and monocular approaches. International Journal of Computer Vision, 74, 343-364.
Ma, X., Li, X., & Qiao, H. (2001). Fuzzy neural network-based real-time self-reaction of mobile robot in unknown environments. Mechatronics, 11, 1039-1052.
Milford, M. J. (2008). Robot Navigation from Nature: Simultaneous Localisation, Mapping, and Path Planning Based on Hippocampal Models. (Kindle Edition ed.) Springer.
Montaner, M.B., & Ramirez-Serrano, A. (1998). Fuzzy knowledge-based controller design for autonomous robot navigation. Expert Systems with Applications, 14, 179-186.
Osamu, H., Masayuki, M., Mei, H. Z., & Kazuharu, K. (2002). A neural network model for encoding and perception of vowel sounds. Neurocomputing, 44-46, 435-442.
Ray, A. K., Behera, L., & Mo, J. (2008). Sonar-Based Rover Navigation for Single or Multiple Platforms: Forward Safe Path and Target Switching Approach. Systems Journal, IEEE, 2, 258-272.
Richard, B. (2009). Cell Assemblies- A new type of neural network. http://richardbowles.tripod.com/ca/intro.htm
Weitzenfeld, A. (2008). From schemas to neural networks: A multi-level modelling approach to biologically-inspired autonomous robotic systems. Robotics and Autonomous Systems, 56, 177-197.
Xiaowei, M., Yun, F. M., Hegao, C., & Khatib, O. (2005). The fuzzy navigator with a local minima solver for real-time self-reaction of a mobile robot in flexible manufacturing cell/system (FMC/FMS). In Systems, Man and Cybernetics, 2005 IEEE International Conference on.
Xu, W. L., Tso, S. K., & Lu, Z. K. (1998). A Virtual Target Approach for Resolving the Limit Cycle Problem. In Victoria, B.C., Canada.
Xu, W. L. & Tso, S. K. (1999). Sensor-based Fuzzy Reactive Navigation of a Mobile Robot through Local Target Switching. IEEE Transactions on Systems, Man, Cybernetics, 29, 451-459.
Zadeh, L. A. (1965). Fuzzy set. Information and Control, 8, 338-353.
Zhu, A. M. & Yang, S. X. (2007). Neurofuzzy-based approach to mobile robot navigation in unknown environments. Ieee Transactions on Systems Man and Cybernetics Part C-Applications and Reviews, 37, 610-621.