別 以嵌入式系統實現人工肌肉氣壓缸之系統識別 以嵌入式系統實現人工肌肉氣壓缸之系統識別 以嵌入式系統實

Abstract

人工肌肉氣壓缸的內部摩擦力，使其具高度非線性特性難以精確控制，故本文以 LuGre 摩擦模型 摩擦模型 摩擦模型 摩擦模型建立人工肌肉氣壓缸的摩擦模型，並以粒子群優化演算法(PSO)求解未知的摩擦參數，以完成人工肌肉氣壓缸的系統識別。

本研究並以可程式系統單晶片PSoC 作為人工肌肉氣壓缸實驗平台的資料擷取卡，藉由感測器所測量得到的訊號得到人工肌肉氣壓缸的壓力與位置，並將取得之訊號資料繪製成人工肌肉氣壓缸的遲滯曲線(Hysteresis curve) 以為摩擦模型識別用。

Pneumatic artificial muscle is a highly nonlinear system because of the internal friction effects, so the LuGre friction model is developed for the friction model of the pneumatic artificial muscle in this paper, and the Particle Swarm Optimization Algorithm (PSO) is used to solve the unknown parameters of the friction model for the system identification of the pneumatic artificial muscle.

In addition, the Programmable System on Chip (PSoC) is used for the data acquisition of the pneumatic artificial muscle system. The data about the displacement and the pressure of pneumatic artificial muscle are measured for the development of the hysteresis loop in the pneumatic artificial muscle.