本論文主要目的是利用灰色系統理論於輪型機器人之避障路徑規劃,機器人兼具對動態與靜態障礙物閃避之自主避障功能。 首先利用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,利用灰色避障控制器來實現機器人閃避障礙物之自主行走能力,達成路徑規劃之目的,並透過攝影機來觀察現場的情況,使用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程式來執行實際的機器人運動控制,研究中利用雷射測距儀得到環境的資料,使用模糊控制器控制機器人到達目標點,並在有障礙物的情況下,設計了灰色避障控制器以完成機器人自主避障運動功能。 The objective of this thesis is to make an obstacle avoidance and path planning robot by grey system theory. The robot is able to avoid dynamic and static obstacle automatically. The center position is defined by the motor encoders of the Segway RMP 50. Then, a fuzzy controller is developed by two input variables (angular and distance errors) to drive the angular displacement and velocity of the Segway. The position sensor is a laser ranger finder in this system. It is installed on the top of the Segway to find out the environmental information for the grey obstacle avoidance controller. The robot is able to make an obstacle avoidance automatically. The computational results are demonstrated by the Matlab and VC++ software program. The experimental results show this intelligent robot works very well.