Modelsko prediktivno upravljanje autonomnim vozilom pri prilazu prometnim semaforima

master's thesis

• Title (english): Model predictive control of autonomous vehicle approaching traffic lights
• Author: Toni Fain
• Mentor: Branimir Škugor (mentor)
• Granter: University of Zagreb; Faculty of Mechanical Engineering and Naval Architecture; Zagreb
• Defense date and country: 2024-07-15, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Traffic and Transport Technology; Intelligent Transport Systems and Logistics

Abstract

Kako tehnologija autonomne vožnje napreduje, optimizacija potrošnje energije postaje ključna,
posebno pri prilasku semaforima gdje zaustavljanje i pokretanje mogu značajno utjecati na
ukupnu energetsku učinkovitost. Ovaj rad fokusira se na optimiranje trajektorije brzine vozila
pri prilasku semaforu s ciljem postizanja udobne vožnje uz minimalnu potrošnju energije.
Korišteno je optimiranje brzine vozila na pomičnom horizontu, poznatije i kao modelsko
prediktivno upravljanje (eng. Model Predictive Control, MPC). Problem maksimiziranja
energetske učinkovitosti može se formulirati kao optimizacijski problem, pod uvjetom da je
dostupna predikcija uvjeta vožnje duž rute. Prediktivne sposobnosti MPC-a čine ga idealnim za
uključivanje predviđanja stanja semafora te prilagođavanje brzine i ubrzanja vozila u skladu s
optimizacijom potrošnje energije. Pretpostavka je da su trenuci budućih promjena stanja
semafora poznati. Simulirani su razni scenariji kako bi se analiziralo ponašanje vozila pri
prilasku semaforima u različitim uvjetima. Ti scenariji uključuju varijacije u trajanju
semaforskih faza i njihovim početnim stanjima kako bi se procijenila učinkovitost predložene
MPC upravljačke strategije.

Abstract (english)

As autonomous driving technology advances, energy consumption optimization becomes
crucial, especially when approaching traffic lights, where stopping and starting can significantly
affect overall energy efficiency. This thesis focuses on optimizing the vehicle speed trajectory
when approaching a traffic light to achieve a comfortable ride with minimal energy
consumption. Vehicle speed profile optimization over a receding horizon, also known as Model
Predictive Control (MPC), is utilized. The problem of maximizing energy efficiency can be
formulated as an optimization problem, provided that a prediction of driving conditions along
the route is available. The predictive capabilities of MPC make it ideal for anticipating changes
at traffic lights and adjusting the vehicle's speed and acceleration in line with energy
consumption optimization. The assumption is that the timing of future traffic light state changes
is known in advance. Various scenarios are simulated to analyze the vehicle's behavior when
approaching traffic lights under different conditions. These scenarios include variations in
traffic light phase durations and their initial states to assess the effectiveness of the MPC
strategy.

• Keywords: autonomna vožnja; modeliranje; optimiranje potrošnje energije; modelsko prediktivno upravljanje (MPC); semafori; udobnost vožnje; simulacija
• Keywords (english): autonomous driving; modeling; energy consumption optimization; model predictive control (MPC); traffic light; comfortable ride; simulation
• Language: croatian
• URN:NBN: urn:nbn:hr:235:856838
• Study programme: Title: Mechanical Engineering; Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra inženjer/inženjerka strojarstva (magistar/magistra inženjer/inženjerka strojarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Open access
• Created on: 2024-07-10 07:23:27