prikaz prve stranice dokumenta Modelsko prediktivno upravljanje za autonomne utrke
No public access
undergraduate thesis
Modelsko prediktivno upravljanje za autonomne utrke
Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing, 2022. urn:nbn:hr:168:519999
Lopotar, Matej
University of Zagreb
Faculty of Electrical Engineering and Computing

Institutional repository: FER Repository

Cite this document

Lopotar, M. (2022). Modelsko prediktivno upravljanje za autonomne utrke (Undergraduate thesis). Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing. Retrieved from https://urn.nsk.hr/urn:nbn:hr:168:519999

Lopotar, Matej. "Modelsko prediktivno upravljanje za autonomne utrke." Undergraduate thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, 2022. https://urn.nsk.hr/urn:nbn:hr:168:519999

Lopotar, Matej. "Modelsko prediktivno upravljanje za autonomne utrke." Undergraduate thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, 2022. https://urn.nsk.hr/urn:nbn:hr:168:519999

Lopotar, M. (2022). 'Modelsko prediktivno upravljanje za autonomne utrke', Undergraduate thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, accessed 17 June 2024, https://urn.nsk.hr/urn:nbn:hr:168:519999

Lopotar M. Modelsko prediktivno upravljanje za autonomne utrke [Undergraduate thesis]. Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing; 2022 [cited 2024 June 17] Available at: https://urn.nsk.hr/urn:nbn:hr:168:519999

M. Lopotar, "Modelsko prediktivno upravljanje za autonomne utrke", Undergraduate thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, Zagreb, 2022. Available at: https://urn.nsk.hr/urn:nbn:hr:168:519999

Metadata
TitleModelsko prediktivno upravljanje za autonomne utrke
Title (english)Model predictive control for autonomous racing
AuthorMatej Lopotar
MentorIvan Petrović (mentor)
Committee memberIvan Petrović (član povjerenstva)
Committee memberBorna Bićanić (član povjerenstva)
Committee memberKarlo Koledić (član povjerenstva)
GranterUniversity of Zagreb
Faculty of Electrical Engineering and Computing
Zagreb
Defense date and country2022-07-06, Croatia
Scientific / art field,
discipline and subdiscipline		TECHNICAL SCIENCES
Computing
Abstract
Cilj projekta je napraviti MPC u autonomnom utrkivanju. Za opis dinamike vozila izabrani su kinematicki i dinami ˇ cki model. Kinemati ˇ ck model kasnije je poboljšan ˇ pa je tako nastao prošireni kinematicki model. MPC smo dizajnirali u ˇ cetiri koraka. ˇ Prvi korak je vožnja pomocu Pure Pursuit Controllera kako bi se skupili podaci. Pure ´ Pursuit je algoritam za pracenje puta koji funkcionira tako da ra ´ cuna kutnu brzinu koja ˇ pokrece automobil prema to ´ cki koju vidi unaprijed. Uzima široke zavoje i ne ulazi ˇ agresivno u njih. Drugi korak je Gaussovim procesom ucenja smanjiti razlike izme ˇ du¯ proširenog kinematickog modela i dinami ˇ ckog modela. Tre ˇ ci korak je dizajniranje ´ MPC-a s ispravljenim i poboljšanim proširenim kinematickim modelom. Vozilo tada ˇ puno agresivnije ulazi u zavoje jer želi proci stazu u što kra ´ cem vremenu. Zadnji, ´ cetvrti korak je ažuriranje Gaussovog procesa nakon vožnje s novostvorenim MPC-om ˇ kako bi dobili konacan rezultat. U procesu su korištena dva automobila s razli ˇ citim ˇ parametrima, kod oba je MPC uspješno dizajniran te oba automobila prolaze krug u najkracem mogu ´ cem vremenu s tim da je automobil kojem su svi parametri pove ´ cani ´ u tome uspješniji.
Abstract (english)
Goal of the project was to design MPC for autonomous racing. To describe vehicle dynamic kinematic and dynamic bicycle models were chosen. Kinematic model was lates updated so it became extended kinematic model. MPC was designed in four steps. First step was driving with Pure Pursuit Controller to gather data. Pure Pursuit is algorithm for path tracking which computes angular velocity that moves car from its current position to some look-ahead point on the track. It is not agressive and it takes wider turns. Second step is learning Gaussian process models to reduce extended kinematic model and dynamic model mismatch. Third step is designing MPC with corrected and updated extended kinematic model. Car is more agressive when turning because it wants to be faster in order to finish the lap in a short time. Fourth step is updating Gaussian process after driving with MPC so we can get final result. Two cars with different parameters were used in the experiment and MPC was successfuly created for both of them so both cars finish their laps in the shortest amount of time. Second car with updated parameters was more successful than first car which did not have its parameters changed.
Keywords
Keywords (english)
Languagecroatian
URN:NBNurn:nbn:hr:168:519999
Study programmeTitle: Computing
Study programme type: university
Study level: undergraduate
Academic / professional title: sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva (sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva)
Type of resourceText
File originBorn digital
Access conditionsClosed access
Terms of useLicence In copyright icon
Public note
Created on2023-01-15 10:41:53