Full Text Available

Note: Clicking the button above will open the full text document at the original institutional repository in a new window.

From simulation to reality : assessing the efficacy of pure pursuit, MPC, and MPCC on the F1Tenth platform

Kong, X. 2025. From Simulation to Reality: Assessing the Efficacy of Pure Pursuit, MPC, and MPCC on the F1Tenth Platform. Unpublished masters thesis. Stellenbosch: Stellenbosch University [online]. Available: https://scholar.sun.ac.za/items/a9c1084b-d0fa-47a3-9229-928d6153fac7

Saved in:
Bibliographic Details
Main Author: Kong, Xiangyiming
Other Authors: Fisher, Callen
Format: Thesis
Published: Stellenbosch : Stellenbosch University 2025
Subjects:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1867613995575279616
access_status_str Open Access
author Kong, Xiangyiming
author2 Fisher, Callen
author_browse Fisher, Callen
Kong, Xiangyiming
author_facet Fisher, Callen
Kong, Xiangyiming
author_sort Kong, Xiangyiming
collection Thesis
dc_rights_str_mv Stellenbosch University
description Kong, X. 2025. From Simulation to Reality: Assessing the Efficacy of Pure Pursuit, MPC, and MPCC on the F1Tenth Platform. Unpublished masters thesis. Stellenbosch: Stellenbosch University [online]. Available: https://scholar.sun.ac.za/items/a9c1084b-d0fa-47a3-9229-928d6153fac7
format Thesis
id oai:scholar.sun.ac.za:10019.1/132590
institution Stellenbosch University (South Africa)
last_indexed 2026-06-10T12:45:00.328Z
license_str Other — see source repository
provenance_str_mv Harvested via OAI-PMH from SUNScholar — Stellenbosch University Repository
publishDate 2025
publishDateRange 2025
publishDateSort 2025
publisher Stellenbosch : Stellenbosch University
publisherStr Stellenbosch : Stellenbosch University
record_format dspace
source_str SUNScholar — Stellenbosch University Repository
spelling oai:scholar.sun.ac.za:10019.1/132590 From simulation to reality : assessing the efficacy of pure pursuit, MPC, and MPCC on the F1Tenth platform Kong, Xiangyiming Fisher, Callen Evans, Benjamin Stellenbosch University. Faculty of Engineering. Dept. of Electrical and Electronic Engineering. Autonomous vehicles -- Control systems Mobile robots -- Automatic control Predictive control Robotics -- Computer simulation UCDT Kong, X. 2025. From Simulation to Reality: Assessing the Efficacy of Pure Pursuit, MPC, and MPCC on the F1Tenth Platform. Unpublished masters thesis. Stellenbosch: Stellenbosch University [online]. Available: https://scholar.sun.ac.za/items/a9c1084b-d0fa-47a3-9229-928d6153fac7 Thesis (MEng)--Stellenbosch University, 2025. ENGLISH ABSTRACT: This study investigates the sim-to-real gap in autonomous racing using the F1Tenth platform. Through literature, three control algorithms based on their computational complexity were selected to be evaluated in simulation and on the physical F1Tenth Radio-Controlled (RC) vehicle. These control algorithms are Pure Pursuit, Model Predictive Control (MPC) and Model Predictive Contouring Control (MPCC). Simulation experiments introduced external disturbances such as noise and delay to the odometry, throttle and steering commands to assess each algorithm’s robustness and performance. Results show that in simulation, MPCC produced the fastest lap time but struggled with throttle disturbances. MPC demonstrated a balanced performance, showing effective tracking of the reference speed and trajectory. Pure Pursuit exceled in its simplicity in its computational demands and robustness to external disturbances with tracks that have minimal change in curvature. However, it performed poorly when the track had rapidly changing curvatures. The physical experiments highlighted the significance of localisation inaccuracies, which stemmed from the particle filter’s reliance on distinct environmental features. A proposed solution incorporating the Kalman filter, real-time delay compensation and adaptive steering control significantly improved the real-world performance, particularly for MPC. However, MPCC’s computational demands on the physical vehicle’s onboard computer limited its real-world performance, suggesting the need for future computational optimisations. AFRIKAANSE OPSOMMING: Hierdie studie ondersoek die sim-tot-werklike gaping in outonome wedrenne deur die F1Tenth-platform te gebruik. Deur middel van literatuur is drie beheeralgoritmes gekies op hul berekeningskompleksiteit om in simulasie en op die fisiese F1Tenth radio-beheer RC voertuig ge¨evalueer te word. Hierdie beheeralgoritmes is Pure Pursuit, Model Predictive Control (MPC) en Model Predictive Contouring Control (MPCC). Simulasie-eksperimente het eksterne steurnisse soos geraas en vertraging by die afstandsmeting, versneller en stuurbevele gevoeg om elke algoritme se robuustheid en werkverrigting te assesseer. Resultate toon dat MPCC in simulasie die vinnigste rondtetyd opgelewer het, maar met versnellerversteurings gesukkel het. MPC het ’n gebalanseerde prestasie getoon, wat effektiewe volging van die verwysingspoed en trajek toon. Pure Pursuit het uitgeblink in sy eenvoud in sy berekeningsvereistes en robuustheid teen eksterne versteurings vir bane wat minimale verandering in kromming het. Dit het egter swak gevaar vir bane met vinnig-veranderende krommings. Die fisiese eksperimente het die belangrikheid van lokaliseringsonakkuraathede uitgelig, wat spruit uit die partikelfilter se afhanklikheid van onderskeibare omgewingskenmerke. ’n Voorgestelde oplossing wat Kalman-filter, intydse vertragingsvergoeding en aanpasbare stuurbeheer insluit, het die werklike werkverrigting aansienlik verbeter, veral vir MPC. MPCC se verwerkingsvereistes het egter sy werklike werkverrigting op die fisiese voertuig se aanboordrekenaar beperk, wat die behoefte aan toekomstige berekeningsoptimisering voorstel. Masters 2025-06-11T10:35:31Z 2025-06-11T10:35:31Z 2025-03 Thesis https://scholar.sun.ac.za/handle/10019.1/132590 Stellenbosch University xiii, 83 pages : illustrations application/pdf Stellenbosch : Stellenbosch University
spellingShingle Autonomous vehicles -- Control systems
Mobile robots -- Automatic control
Predictive control
Robotics -- Computer simulation
UCDT
Kong, Xiangyiming
From simulation to reality : assessing the efficacy of pure pursuit, MPC, and MPCC on the F1Tenth platform
title From simulation to reality : assessing the efficacy of pure pursuit, MPC, and MPCC on the F1Tenth platform
title_full From simulation to reality : assessing the efficacy of pure pursuit, MPC, and MPCC on the F1Tenth platform
title_fullStr From simulation to reality : assessing the efficacy of pure pursuit, MPC, and MPCC on the F1Tenth platform
title_full_unstemmed From simulation to reality : assessing the efficacy of pure pursuit, MPC, and MPCC on the F1Tenth platform
title_short From simulation to reality : assessing the efficacy of pure pursuit, MPC, and MPCC on the F1Tenth platform
title_sort from simulation to reality assessing the efficacy of pure pursuit mpc and mpcc on the f1tenth platform
topic Autonomous vehicles -- Control systems
Mobile robots -- Automatic control
Predictive control
Robotics -- Computer simulation
UCDT
url https://scholar.sun.ac.za/handle/10019.1/132590
work_keys_str_mv AT kongxiangyiming fromsimulationtorealityassessingtheefficacyofpurepursuitmpcandmpcconthef1tenthplatform