DocumentCode :
3337421
Title :
Identification of the 2-Axes Pneumatic Artificial Muscle (PAM) Robot Arm Using Double NARX Fuzzy Model and Genetic Algorithm
Author :
Anh, Ho Pham Huy ; Ahn, Kyoung Kwan ; Yoon, Jong Il
Author_Institution :
Sch. of Mech. & Automotive Eng., Univ. of Ulsan, Ulsan
fYear :
2008
fDate :
9-11 April 2008
Firstpage :
84
Lastpage :
89
Abstract :
In this paper, a novel Forward Double NARX fuzzy model is used for modeling and identifying simultaneously both joints of the prototype 2-axes PAM robot arm. The highly nonlinear coupling feature of both links of the 2-axes PAM robot arm is modeled thoroughly through a Forward Double NARX Fuzzy Model-based identification process using experimental input-output training data. The evaluation of different types of Double NARX Fuzzy Model of the 2-axes PAM robot arm with various ARX model structure will be discussed. For first time, the nonlinear Forward Double NARX Fuzzy Model scheme of the prototype 2-axes PAM robot arm has been investigated. The results show that the nonlinear forward Double NARX Fuzzy Model trained by genetic algorithm yields more performance and higher accuracy than the traditional Fuzzy model.
Keywords :
fuzzy set theory; genetic algorithms; identification; manipulators; fuzzy model; genetic algorithm; identification process; nonlinear forward Double NARX Fuzzy Model; pneumatic artificial muscle robot arm; Automotive engineering; Biological cells; Electronic mail; Genetic algorithms; Genetic mutations; Intelligent robots; Manipulator dynamics; Muscles; Neural networks; Prototypes; 2-axes PAM robot arm; genetic algorithm; modeling and identification; nonlinear Forward Double NARX Fuzzy model; pneumatic artificial muscle (PAM);
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Smart Manufacturing Application, 2008. ICSMA 2008. International Conference on
Conference_Location :
Gyeonggi-do
Print_ISBN :
978-89-950038-8-6
Electronic_ISBN :
978-89-962150-0-4
Type :
conf
DOI :
10.1109/ICSMA.2008.4505618
Filename :
4505618
Link To Document :
بازگشت