本論文將探討模糊控制及灰色預測應用於三軸機械手臂的運動控制。最近,模糊控制器已成新興的潮流,雖然其設計時很方便,但對機械手臂的設計,往往會有超越量的情形發生及較差的暫態和穩態響應。因此,針對這個問題,本論文將採取灰色預測當做前饋式補償器,如此設計的模糊控制器,當超越量產生之前,便以力量之補償方式降下來,可預防超越量的產生,並且於上升時間也有不錯的暫態響應及較小的穩態誤差,並且使用最佳化之演算法-基因演算法去調適灰色預測內重要參數ak,使整個系統設計更加完善。再進一步討論,將設計比例-微分-積分控制器,並且與提出的方法做比較及分析。此外,本論文使用模糊速度控制器去控制手臂的等速運動及防止干擾(如負載量的變化)。最後,發現在大部分實驗中,機械手臂均可達到預期目標,而且藉由視窗化的軟體,將所有實驗及設計均視窗化,以求達到良好的人機使用介面。本論文將在最後探討小部分結果不甚理想的原因,並提出未來可以改進的方向。 In this research, we first provide a fuzzy controller with the grey prediction feedforward compensator to replace the conventional position controller. The grey model has some efficacy in forecasting the overshoot of the response. In order to obtain the accurate prediction, the constant parameter ak of grey model needs to be optimal. Hence, we use a SGA (Simple Genetic Algorithm) to obtain the proper constant ak. Next, the fuzzy controller is presented to solve the speed control problem. The fuzzy speed controller not only maintains the constant speed but also prevents the disturbance. In the implementation process, the design is completed under the windows software. Therefore, it is easy to fix the parameters rapidly and get the fixed sampling time. Then, the discussion of the experimental results is given. Finally, it is found that the performance of both the position control and the speed control of a three-axis robot arm is satisfactory.