本論文實現一個利用雙手運動來補償人形機器人重心移動的平衡控制方法。本論文透過一台小型人形機器人之腳底加裝壓力感測器來量測重心,再加上本論文所提出之雙手運動補償方法來讓人形機器人可以實現雙腳及單腳支撐下的自主平衡,並完成如抬腳或踢球等需要平衡能力的動作。在機器人的動作軌跡部分,本論文運用工業用核心電腦計算出機器人雙足末端點位置後傳給FPGA (Field-Programmable Gate Array),再透過FPGA內部之模組運算相對於末端點的關節馬達角度以控制人形機器人的姿態。在平衡的部分,本論文以機器人的重心位置來判斷機器人的平衡狀態,而後根據所提出的模糊控制器來修正機器人的雙手轉動以完成平衡動作。由實驗結果得知,本論文所提出之系統可以有效的修正機器人姿態,以增加平衡穩定性。
A humanoid robot balance control method based on both of the hand movement is proposed in this thesis. A humanoid robot with 23 degree of freedom and 8 force sensors is designed as the platform. The proposed balance method determines the movement of the hands to compensate the movement of the robot’s center of gravity. The robot is able to execute the lifting leg or kicking ball motion in the one leg support situation. An IPC is applied to calculate the trajectory of the foot of the humanoid robot. The result of the trajectory of the foot is sent to FPGA (Field-Programmable Gate Array) board for calculate the relate angle of the motor based on inverse kinematic. The proposed system measures the center of gravity of the humanoid robot by the force sensors. The proposed fuzzy controller calculates the hand movement to balance the robot by the center of the gravity of the robot. The practical experiments with different situations are presented to illustrate the efficiency of the proposed method.