Simulation of the 6-DOF model of Discrete State Manipulators (DSMs) Based on Neuro-Fuzzy Architecture

Pasila, Felix and Alimin, Roche and Tanoto, Yusak (2016) Simulation of the 6-DOF model of Discrete State Manipulators (DSMs) Based on Neuro-Fuzzy Architecture. In: The International Conference on Information Technology and Digital Applications (ICITDA), 08-11-2016 - 08-11-2016, Yogjakarta - .

	PDF Download (986Kb)
Preview	PDF (paper - Roche) Download (2066Kb) | Preview
Preview	PDF (cek plagiasi - Roche) Download (1604Kb) | Preview

Abstract

This paper reports the simulation of neuro-fuzzy architecture of the discrete state manipulators (DSMs) with 6 degree of freedom (6-DOF) in Matlab environment. The DSMs are special kind of discrete manipulator with massive pneumatic actuators that can be switched among limited number of discrete states. We introduce three-state DSMs model, called as ternary DSMs (t-DSMs), which is driven by discrete forces and have continuous motions. The main problem of the DSMs are how to control such as the position of the manipulator by choosing the best combination of actuators states which called as inverse static problem (ISP) of DSMs. This paper proposed an architecture which is based on the Neuro-Fuzzy Takagi Sugeno (NFTS) inference scheme with Gaussian membership functions as ISP solution of the proposed DSMs. The training algorithm needs at least one million iterations with different membership functions, employ 25.9% of the input-output correspondences dataset from the known input and output. For training database, the NFTS model generates 189 dataset from the 729 possible dataset. After one week training for searching optimized parameters within several membership function (M), the validation testing found the best error in 1.51% using M = 9.

Actions (login required)