EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Comparative Modeling and Experimental Validation of Two Four-Wheel Omnidirectional Locomotion Architectures for a Modular Mobile Robot

Iosif-Adrian Maroșan, Alexandru Bârsan, George Constantin, Sever-Gabriel Racz, Radu-Eugen Breaz, Claudia-Emilia Gîrjob, Mihai Crenganiș, Cristina-Maria Biriș

Applied Sciences · 08.04.2026 · 16 (8) · 3646

DOI: 10.3390/app16083646 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Applied Sciences, online erschienen am 08.04.2026. Verfasst von Iosif-Adrian Maroșan und 7 weiteren Autor:innen. Die Arbeit nennt 33 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Lucian Blaga University of Sibiu through the research grant LBUS-IRG-2024.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

This paper presents a comparative modeling and experimental validation study for a modular four-wheel omnidirectional mobile robot, focusing on two locomotion architectures implemented on the same platform: four omni wheels (90° rollers) and four Mecanum wheels (45° rollers). Both configurations were evaluated under identical benchmark conditions on a 1 m × 1 m square path (4 m total path length), using the same nominal 12 V supply and the same test duration, in order to ensure a fair and reproducible cross-architecture comparison. A MATLAB/Simulink–Simscape dynamic model was developed for both architectures, while experimental validation was performed using Hall-effect current sensors integrated into the drive modules. Based on the measured and simulated motor currents, a 12 V-based electrical input-power estimate was evaluated at both motor and robot level. For the considered benchmark, the four-Mecanum configuration exhibited a lower measured input-power estimate than the four-omni configuration (17.88 W vs. 25.75 W), corresponding to an approximate reduction of 30.6% under the adopted assumptions. At robot level, the deviation between simulated and measured total input-power estimate was 3.70% for the four-omni architecture and 21.42% for the four-Mecanum architecture, indicating higher predictive agreement for the omni-wheel model in its present form. The comparative analysis also suggests that wheel–ground interaction and roller geometry influence not only the measured current demand but also the level of agreement between simulation and experiment. Although the present study is limited to a single standardized benchmark and nominal-voltage conditions, it provides a controlled basis for comparing the two locomotion solutions and for identifying directions for further model refinement. The findings should therefore be interpreted as benchmark-specific comparative results offering practical guidance for locomotion architecture selection and for future refinement of friction-aware omnidirectional robot models.

Förderung

  • Lucian Blaga University of Sibiu through the research grant LBUS-IRG-2024

Zitieren

APA

Maroșan, I., Bârsan, A., Constantin, G., Racz, S., Breaz, R., Gîrjob, C., Crenganiș, M., & Biriș, C. (2026). Comparative Modeling and Experimental Validation of Two Four-Wheel Omnidirectional Locomotion Architectures for a Modular Mobile Robot. Applied Sciences, 16(8), 3646. https://doi.org/10.3390/app16083646

Vancouver

Maroșan I, Bârsan A, Constantin G, Racz S, Breaz R, Gîrjob C, et al. Comparative Modeling and Experimental Validation of Two Four-Wheel Omnidirectional Locomotion Architectures for a Modular Mobile Robot. Applied Sciences. 2026;16(8):3646. doi:10.3390/app16083646

BibTeX
@article{Maroan2026,
  author = {Maroșan, Iosif-Adrian and Bârsan, Alexandru and Constantin, George and Racz, Sever-Gabriel and Breaz, Radu-Eugen and Gîrjob, Claudia-Emilia and Crenganiș, Mihai and Biriș, Cristina-Maria},
  title = {Comparative Modeling and Experimental Validation of Two Four-Wheel Omnidirectional Locomotion Architectures for a Modular Mobile Robot},
  journal = {Applied Sciences},
  year = {2026},
  volume = {16},
  number = {8},
  pages = {3646},
  doi = {10.3390/app16083646}
}
Kostenlos starten →