PERANCANGAN PENGENDALI ROBOT BERGERAK BERBASIS PERILAKU MENGGUNAKAN PARTICLE SWARM FUZZY CONTROLLER

Andi Adriansyah

Abstract


Paper ini memaparkan perancangan pengendali robot berbasis perilaku menggunakan Fuzzy, di mana parameter Fuzzy ditala secara otomatis menggunakan Particle Swarm Optimization (PSO) yang diistilahkan dengan Particle Swarm Fuzzy Controller (PSFC). Suatu fungsi tertentu dirancang untuk meningkatkan performa proses pencarian PSO. Fungsi tersebut mengubah harga bobot inersia menjadi berkurang secara sigmoid (Sigmoid Decreasing Inertia Weight). Empat buah perilaku robot dirancang menggunakan PSFC. Kemudian seluruh perilaku tersebut juga dikoordinasikan menggunakan PSFC. Beberapa simulasi pengendalian pergerakan robot dan percobaan dengan robot MagellanPro telah dilakukan untuk menguji performa algoritma yang dirancang. Algoritma lain, Genetic Fuzzy Controller (GFC) digunakan sebagai pembanding. Dari hasil pengujian dapat dikatakan bahwa pengendali yang dirancang memiliki kemampuan yang baik untuk menyelesaikan tugasnya pada suatu lingkungan nyata.

This paper describes the design of robots controllers based on behaviour using Fuzzy, in which the Fuzzy parameters are automatically tuned using the Particle Swarm Optimization (PSO) which is termed the Particle Swarm Fuzzy Controller (PSFC). A particular function is designed to improve the performance of PSO search process. That particular function changes the value of the inertia weight, so it’s decreased in sigmoid (Sigmoid Decreasing Inertia Weight). Four types of robots behaviour are designed and coordinated using the PSFC. Some simulation of the robot movement control and experiments with the robot MagellanPro have been conducted to test the performance of the algorithm that have been designed. Another algorithm, Genetic Fuzzy Controller (GFC) is used as a comparison. From the test results, it can be said that the controllers that have been designed, have a good ability to accomplish its task in a real environment.

Keywords


behavior-based robots; fuzzy logic controllers; moving robot; particle swarm optimization; pengendali logika fuzzy; robot berbasis perilaku; robot bergerak

Full Text:

PDF


DOI: http://dx.doi.org/10.21609/jiki.v3i1.135

Refbacks

  • There are currently no refbacks.


Copyright © Jurnal Ilmu Komputer dan Informasi. Faculty of Computer Science Universitas Indonesia.

Creative Commons License

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.

View JIKI Statistic