本論文主要目的是設計及製作一個由三個直流伺服馬達作為關節，長短左右對稱的四節機器人，其外側連桿較短，內側連桿較長，來探討四節機器人在未知載重下的垂直站立動態平衡運動控制。首先利用個人電腦(PC)設計四節機器人的垂直平衡站立運動路徑規劃，再利用類神經網路演算法與模糊控制完成垂直平衡站立運動控制的目標，硬體方面包含了機構設計、直流伺服馬達、傾斜計、數位訊號處理器(DSP)、驅動器的整合。在垂直平衡站立運動中，機器人之重心是判斷平衡與否的重要因素，利用各個馬達之目標角度及速度，配合類神經網路演算法訓練，讓系統在失敗中學習將機構抬起達到機器人穩定站立狀態，當機器人在穩定狀態下，再使用重心法配合模糊控制模擬出四節機器人在未知載重下的垂直平衡站立。最後，模擬出實驗結果印證了機器人在未知載重狀態下的自我學習能力。 The objective of this thesis is to develop a four-link robot to stand up vertically and stably with an unknown loading on a linkage. There are three joints and four links in the robot system. The outside links are long, and inside links are short. The joints are driven by three DC server motors respectively.The sequence of the standing behaviors is assigned by the designed program in a PC. The artificial neural network (ANN) and fuzzy theory are applied to design the controller for the robot. The hardware of this system includes mechanism, DC server motors, a tilt sensor, digital signal process (DSP) board, and motor drivers. The center of gravity (COG) of the robot is an important factor in the standing process. The robot is trained to stand up by angles and angular velocity of motors by using the ANN algorithm. Then, the COG of the robot is controlled to stand up vertically and stably by the fuzzy control theory. The simulational and experimental results indicate that the four-link robot is able to stand up by itself with an unknown loading.
