Ibadi, Iksan (2023) Implementasi Genetic Algorithm untuk Optimisasi Parameter Model Predictive Controller pada Sistem Pengaturan Kemudi di Mobil Otonom. Other thesis, Institut Teknologi Sepuluh Nopember.
Text
07111840000026-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 April 2025. Download (2MB) | Request a copy |
Abstract
Algoritma Model Predictive Controller (MPC) seringkali digunakan sebagai kontroler utama pada sebuah sistem kemudi mobil otonom, salah satunya yakni digunakan pada dinamika path-tracking mobil otonom. Dinamika path-tracking mobil otonom memiliki model sistem yang nonlinier dan memiliki banyak state variabel, sehingga MPC lebih sering digunakan dibandingkan dengan kontroler jenis lainnya. Akan tetapi terjadi permasalahan pada proses desain MPC, yakni terkait penentuan parameter yang digunakan yakni horizon prediksi dan horizon kontrol yang seringkali ditentukan berdasarkan trial-and-error sehingga menghasilkan performa sistem yang kurang optimal. Pada penelitian ini, Genetic Algorithm (GA) akan diimplementasikan untuk mencari nilai parameter MPC yang optimal dengan cara mengoptimisasi nilai error, tingkat kenyamanan berkendara, sinyal kontrol yang dikeluarkan kontroler serta lamanya waktu komputasi. Dari percobaan yang dilakukan untuk satu skenario menunjukkan bahwa metode yang digunakan efektif dalam meningkatkan performa sistem total apabila dibandingkan dengan proses optimisasi yang memprioritaskan satu komponen optimisasi saja. Dari proses optimisasi GA yang telah didesain berdasarkan skenario yang ditentukan, didapatkan pasangan horizon optimal (11,6) yang menghasilkan nilai total cost function 0.0394 untuk waktu simulasi 61.2s menghasilkan skor IAE 856.659.
==============================================================================================================================
Model predictive controller (MPC) algorithm are often used for the main controller in autonomous vehicle’s steering control, especially for path-tracking processes. MPC is chosen in most cases in autonomous vehicle control system because the models of the system is nonlinear system, and sometimes it has many state variables and constraints that makes MPC shines compared to other controller algorithm. But there is a problem when designing an MPC. The parameters in MPC, prediction and control horizon, are chosen manually by the designers by using trial-and-error methods, and sometimes resulting in suboptimal system performances. In this paper, a new method based on Genetic Algorithm (GA) is proposed to achieve best prediction and control horizon considering minimization of several terms such as lateral error, driving convenience, energy for generating control signal and computational burdens. The simulation results show the effectiveness of the proposed method in terms of overall performances compared to optimization methods that only prioritized one optimization component. The proposed method for determined scenario resulting in optimal horizons (11,6) with cost value 0.0394 and IAE score 856.659 for 61.s simulation time.
Item Type: | Thesis (Other) |
---|---|
Uncontrolled Keywords: | mobil otonom, pengaturan kemudi, optimisasi, Model Predictive Control, autonomous car, steering control, optimization, Genetic Algorithm |
Subjects: | T Technology > TJ Mechanical engineering and machinery > TJ217.6 Predictive Control |
Divisions: | Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Electrical Engineering > 20201-(S1) Undergraduate Thesis |
Depositing User: | Iksan Ibadi |
Date Deposited: | 03 Feb 2023 19:51 |
Last Modified: | 03 Feb 2023 19:51 |
URI: | http://repository.its.ac.id/id/eprint/96145 |
Actions (login required)
View Item |