在無線感測器網路 (Wireless Sensor Networks)中，環境監視(surveillance)是一個很重要的應用。為了在廣大且艱難的監測區域中提供環境監視的服務，使用機器人代替人工佈置感測器(sensors)，將有助於監測區域的範圍擴大到人類無法到達的區域，並增加佈置感測器上方便性及準確性。在這篇論文中，我們提出了利用避障以及偏移得知的機器人佈點演算法，來佈置無線感測器網路(Obstacle-Free and Bias-Aware Robot Deployment Algorithm)，機器人能在適當的距離內放下感測器，使得任兩個感測器之間的感測範圍重疊程度最低，但卻能達到充分涵蓋整個監測區域的目的。在佈置感測器的過程中，我們亦將機器人因機械特性所產生的移動偏差予以自我調整，並克服障礙物對佈置感測器之影響。此外，在佈建網路的過程中，亦能使已佈建的感測器盡量進入休眠狀態進行省電，以便增加無線感測器網路的存活時間(lifetime)。由實驗結果顯示出，我們的演算法能以最少的感測器達到最大的監測範圍，並且有較佳的避障效果。 This thesis proposed a robot deployment algorithm that overcomes unpredicted obstacle, corrects bias movement, and meets the full-coverage requirement. The robot will deploy sensors to minimize the overlapping area of sensors’ coverage and achieve the full-coverage purpose with fewer deployed sensors. A spiral movement rule is applied to reduce the number of deployed sensors that requires assisting the robot deployment, resulting most deployed sensors staying in sleep mode during the deployment process. The developed deployment algorithm also takes into consideration the bias movement and unpredicted obstacles, reducing their impacts on full-coverage deployment. Simulation results show that the proposed algorithm can significantly reduce the total number of deployed stationary sensors, achieve the full-coverage purpose, and saves energy consumptions. The developed robot deployment algorithm also works well as unpredicted obstacles encountered or bias movement occurred.
