透過您的圖書館登入
IP:34.230.35.103
  • 學位論文

微型仿生體之氣味追蹤演算法與氣味擴散環境建立

Odor Tracking Algorithm and Scent Distribution Model Development for Micro Bio-mimetic Robot

指導教授 : 陳永耀
若您是本文的作者,可授權文章由華藝線上圖書館中協助推廣。

摘要


微型機械具有感測周圍環境的刺激以及產生相對應的行為,可用在無人監督下幫忙人類完成許多工作,包括許多重複性以及危險性的工作如救災、搜尋有毒氣味來源,以及應用於生產線增加生產力等等。其中,建立一套感測系統能讓微型仿生機械有能力去尋找及定位一個或多個氣味來源是尤其吸引我們作相關及深入的研究在這氣體感測的領域裡。因為此類應用被高度需求在瓦斯漏氣偵測、氣體污染源的追蹤以及救災救難中失火的起始點。 本論文發展了一套氣味追蹤演算法可以讓帶著氣味感測器的微型仿生機械有能力去搜尋及定位氣味來源;此演算法主要部分是基於一種概念”氣味導引”也就是利用單一感測器可以指引微型仿生機械到氣味來源的附近領域。而本論文依此提出一種新的氣味來源定位演算法,讓微型仿生機械可以在離氣味來源較遠的區域先行預測離自己近的氣味來源的位置,當微型機械作一個固定模式的運動時,將氣味感測器感應的數值帶入氣味直接擴散的模型;因此將兩者做類似擬合動作,可以先行預測氣味來源的方位。 總結本論文主要研究在探討有限感測能力下,仿生昆蟲之行為反應模式設計及所能達成之預期工作目標,研究重點在於感測系統之建立及智慧型感測訊號處理與行為反應。本論文主要的研究內容包含兩主題,分別為氣體擴散環境的建立以及氣味追蹤演算法在氣味來源定位的應用。在本論文中,我們為氣味來源定位問題提供詳細的解析,從一般的搜尋方式到建立一套有系統有效率的氣味搜尋演算法;在這些研究方法裡,仿生生物行為的靈感激發給了我們演算法很大的助益。

並列摘要


This thesis presents a methodology for developing a search algorithm which directs self-organized and micro-autonomous robotic systems. Then it demonstrates how this algorithm can be applied to the problem of finding one or more than one odor sources in the indoor environment without constant airflow. The search algorithm is applicable to other task domains and the resulting odor localization system can extend the development of a micro-robot. Specifically, this thesis analyzes a basic collective search task for random and coordinated scent search. It also investigates a set of biologically inspired behaviors that permit a micro-robot to traverse an odor distribution environment to its source and describes the common characteristics of successful algorithms. Collective search and zigzag search are then combined (along with egocentric source declaration) into the full odor localization task which is optimized in simulation. Then, following the design methodology, an odor distribution model with obstacles is presented which is used in simulation to numerically verify the scent search algorithm. Finally, a search behavior is developed for a micro-robotic scent tracking vehicle to collectively “sniff out” locations of high scent concentrations in unknown, geometrically complex environments. The micro-vehicle is assumed to be programmed with guidance and control algorithms. This algorithm is comprised of a sensory compensation sub-algorithm using point source cancellation, a guidance sub-algorithm using spiral surge tracking, zigzag collective search and gradient descent tracking. The concepts of zigzag collective search and point source cancellation are modern concepts introduced within. Simulation results for micro-vehicle are given.

參考文獻


[1] R.A. Russell, D. Thiel, R. Deveza, A. Mackey-Sim, A robotic system to locate hazardous chemical leaks, Proc. IEEE Int. Conf. Robotics and Automation, Nagoya, Japan, May 1995, pp. 556-561.
[2] T. Nakamoto, H. Ishida, T. Moriizumi, An odor compass of localizing an odor source, Sensors and Actuators B 35 ,1996,pp.32-36.
[3] B. Webb. View from the boundary. Biological Bulletin, 200:184, April 2001.
the factory. Red Herring, September 1998.
[6] J. F. Engelberger. Robotics in practice : management and applications of industrial robots. AMACOM, New York, 1980.

被引用紀錄


Tien, C. L. (2006). 多群組微型仿生體之合作型搜尋演算法發展與行為設計 [master's thesis, National Taiwan University]. Airiti Library. https://doi.org/10.6342%2fNTU.2006.10521

延伸閱讀