Prediktivni sustav za održavanje vozila unutar prometne trake

master's thesis

• Title (english): Predictive lane keeping system
• Author: Petar Makarun
• Mentor: Šandor Ileš (mentor)
• Committee member: Šandor Ileš (predsjednik povjerenstva)
• Committee member: Šandor Ileš (predsjednik povjerenstva)
• Committee member: Jadranko Matuško (član povjerenstva)
• Committee member: Jadranko Matuško (član povjerenstva)
• Committee member: Martina Kutija (član povjerenstva)
• Committee member: Martina Kutija (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2020-02-28, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Electrical Engineering

Abstract

U radu je razvijen modelsko prediktivni algoritam upravljanja vozilom koji osigurava održavanje vozila unutar definirane prometne trake. Upravljački algoritam je ispitan u MATLAB/Simulink simulacijskom okruženju na CarSim modelu vozila uz YALMIP alat za modeliranje i OSQP alat za rješavanje optimizacijskog problema. Željena trajektorija vozila tijekom simulacije je određena pomoću senzora za detekciju linija u CarSim alatu. Upravljački algoritam je također ispitan i eksperimentalno na laboratorijskom skaliranom modelu vozila. Željena trajektorija tijekom eksperimenta određena je pomoću računalnog vida. Obradom slike prepoznata je sredina prometne trake, te horizontalni profil prometne trake. Upravljački algoritam na modelu vozila implementiran je pomoću JuMP modelskog jezika za matematičku optimizaciju ugrađenog u Julia programskom jeziku te riješen koristeći OSQP alat za rješavanje kvadratnih optimizacijskih problema.

Abstract (english)

In this thesis, a model predictive control algorithm has been presented for lane keeping system on an autonomous vehicle. The control algorithm was tested in the MATLAB/Simulink simulation environment on the CarSim vehicle model with the YALMIP modeling and optimization tool and OSQP quadratic programming solver. The desired trajectory of the vehicle during the simulation was determined by a lane detection sensor in CarSim. The control algorithm was also tested experimentally on a laboratory scaled model of the vehicle. The desired trajectory during the experiment was determined by computer vision. Using image processing the middle of the driving lane and the horizontal profile of the driving lane was recognized. Control algorithm on the vehicle model was implemented using JuMP modeling language for mathematical optimization embedded in the Julia Programming Language and solved using OSQP quadratic programming solver.

• Keywords: Model prediktivno upravljanje; računalni vid; sustav za zadržavanje vozne trake; YALMIP; Julia; OSQP; MATLAB; CarSim
• Keywords (english): Model predictive control; computer vision; lane keeping system; YALMIP; Julia; OSQP; MATLAB; CarSim
• Language: croatian
• URN:NBN: urn:nbn:hr:168:512970
• Study programme: Title: Electrical Engineering and Information Technology; Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra inženjer/inženjerka elektrotehnike i informacijske tehnologije (magistar/magistra inženjer/inženjerka elektrotehnike i informacijske tehnologije)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2023-01-13 10:19:42