- 學位論文
移動機器人搜尋輻射源策略
Search Strategy of a Mobile Robot for Radiation sources
摘要
本研究提出一個在國際入境場所環境中使用具有裝置輻射偵檢器(閃爍體計數器)的移動式機器人搜尋輻射源的方法。因為國際入境場所的建築物或物品擺設位置無固定，而無從得知真實環境狀況，我們的目標是使移動機器人能夠在未知的複雜環境中找輻射源的分佈，在這整理出兩個問題討論：(1)移動機器人如何在未知環境中避開障礙以及靠近輻射源，(2)估測出輻射源強度及位置。估測輻射源的方法使用粒子濾波器(Particle Filter)進行位置以及強度估測，在估測過程中具有感測網路概念，多樣性感測節點分析各粒子的權重值，使用K-Means分群演算法選擇具有代表性的感測節點。使移動機器人的避障能力和逼近輻射源位置則用人工位能場(Potential Field)的觀念改變演算法。最後實驗移動式機器人在輻射環境空間中偵測，每次移動到下個節點都會記錄此位置的輻射值，將這些輻射資訊分析，判斷環境的實際輻射分佈。並且本研究使用Generalized variance計算粒子的收斂程度，分配慣性與粒子預估吸引向量的權重值，使得搜尋路徑上減省多餘的路徑，增加搜尋輻射源效率。
並列摘要
Searching radiation sources in a nuclear environment plays an important role in securing nuclear safety. In this paper, we present a search strategy of a mobile robot for radiation sources in an unknown environment. The proposed method enables a mobile robot to find radiation sources autonomously because it is harmless to expose the robot to an environment contaminated by radiation. Previous work employed a robot to find radiation sources with a given environmental map. Realistically, once a nuclear accident occurs, the environment would be partially destroyed. Thus, the purpose of this paper aims at navigating a mobile robot to radiation sources without a given environmental map. To achieve the goal, two sub-problems are discussed in this paper: (1) how does a mobile robot navigate itself to radiation sources without a map, and (2) how is the position and intensity of a radiation source estimated by a Geiger counter. For these two problems, an artificial potential field (APF) is adopted to navigate the robot to radiation sources and the direction and intensity of the attractive force in the APF is estimated by a particle filter. We provide a simulation validation of the proposed method which demonstrated good results to find the radiation sources without the map.