下肢外骨骼機器人輔助人體跨步平衡控制

Abstract

本論文「下肢外骨骼機器人輔助人體跨步平衡控制」旨在設計一下肢外骨骼機器人在人體受到較大外力撞擊時能輔助人體跨出步伐以維持平衡。本文利用足底壓力感測器量測人體之足底壓力中心(Center of Pressure, CoP)位置及大小，並以此作為是否進行跨步平衡之依據。安裝於下肢外骨骼機器人之感測器及編碼器則用來計算外骨骼之姿態，並藉由線性倒單擺模型推導出能使人體平衡之跨步位置捕獲點(Capture Point, CP)，再透過人工勢場(Artificial potential field)設計運動軌跡及使用逆運動學將末端點運動軌跡轉換為各關節之角度。控制器以計算出之角度與實際角度間之誤差作為控制輸入，使用PID控制及滑模模糊控制(Fuzzy Sliding Mode Control, FSMC)並探討其效果，最後穿戴於正常人與輕度帕金森氏症患者並輔助其受到外力干擾後的跨步平衡。

The main purpose of this thesis “Stepping Balance Control through Assistance of a Lower-Limb Exoskeleton Robot” aims to design a lower-limb exoskeleton robot that can assist the human body to take a step for maintaining balance when hitting with a large external force. In this thesis, the plantar pressure sensor is used to measure the position and size of the Center of Pressure (CoP) of the human body, which is used as the basis for taking step to maintain balance. Sensors and Encoders on the lower-limb exoskeleton robot are used to calculate the attitude of the exoskeleton. Linear Inverted Pendulum Model is used to derive the Capture Point (CP), and design a trajectory through Artificial Potential Field, then uses Inverse Kinematics to convert the design trajectory into the angle of each joint. The PID control and Fuzzy Sliding Mode Control (FSMC) are selected as controllers and dicuss its effect. Finally, normal people and patients with Parkinson's disease wears the lower-limb exoskeleton to assist their stepping balance after being disturbed by external forces.