本論文主要目的是設計一部能夠在未知環境下進行定位與建立該區域地圖之輪型機器人,同時兼具對動態與靜態障礙物閃避之自主運動功能。首先設計Segway移動控制平台軟體,並利用馬達的編碼器資料定義Segway自我位置,並設計模糊控制器透過相對角度與距離誤差的參數,驅動Segway的馬達角位移及角速度達到控制的目的。 其次運用雷射測距儀透過TCP/IP的傳輸方式與Segway人機介面連結,分析雷射測距儀的資料得知環境資訊,做為Segway移動平台偵測障礙物依據,利用模糊控制器來實現機器人閃避障礙物之自主行走能力,透過Matlab模擬程式來檢視機器人移動軌跡與驗證系統的穩定性,進一步將模擬程式轉換成Visual C++ MFC程式來觸動實際的機器人運作,研究中利用ICP(Iterative Closest Point)演算法將環境資料做匹配,透過兩筆資料的比較可以獲得該兩筆資料之位置差異,包括平移矩陣以及旋轉矩陣,經由反覆的運算後可以估測雷射測距儀的位置,並建立環境地圖,經實驗驗證,有效的將誤差收斂趨近於零,與模擬結果一致,証明本論文所開發之系統確實可行。
The objective of this thesis is to design a robot to have independent motion behaviors and obstacle avoidance function. The robot is also able to make the SLAM(Simultaneous Localization And Mapping) at the unknown environment. First, the software of the robot platform is designed to define the robot position by the motor encoders. Then , the obstacle avoidance controller is developed by two input variables (angular and distance errors) of the fuzzy theory. A laser ranger finder is linked to the segway by the TCP/IP (Transmission Control Protocol / Internet Protocol). The environmental information of laser ranger finder is the important input variables of the obstacle avoidance controller. The simulated results of this system is implemented by the Matlab software. After that, the simulated program is transformed into VC++ MFC program to demonstrate the actual motion behavior of this robot system. The calculation algorithms of ICP(Iterative Closest Point) is applied to check the position error of the environmental data to obtain the estimated position of the laser range finder. The estimated position data are used to calculated the final function of SLAM. Finally, both the simulated and experimental results show that the system work very well.