Comparative of a real-time vision system under Xenomai and PREEMPT_RT

Keywords: Computer vision, real-time vision systems, real-time operating systems

Abstract

This article presents the design and programming of a vision system on two real-time operating systems. The above is for the purpose of making a comparison of their performance. The operating systems used were Xenomai and Linux with the PREEMPT_RT patch. The structure of the vision system in both cases is divided into three threads: the first one in charge of acquiring and processing the images, the second one that displays the images on the screen and the last one for the data transfer (and subsequent analysis out of line). The objective of this proposed structure is the optimization of the execution times; which may be smaller compared to using a single process or thread for the entire vision system. Finally, experiments are described emphasizing the periodicity achieved by performing an image processing corresponding to the calculation of a centroid that gives information about the movement of a simple pendulum.

Downloads

Download data is not yet available.

References

Adam, G. (2021). Real-Time Performance and Response Latency Measurements of Linux Kernels on Single-Board Computers. Computers, vol. 10. p. 64.

Brown, J., Martin, B. (2010). How fast is fast enough? Choosing between Xenomai and Linux for real-time applications. Twelfth Real-Time Linux Workshop, p. 1- 17.

Bugarin, E., Aguilar-Bustos, A. Y. (2014). Control visual para la formación de robots móviles tipo uniciclo bajo el esquema líder-seguidor. Ingeniería, Investigación y Tecnología. XV(4): 593-602.

Burns, A., Wellings, A. (2002). Sistemas de Tiempo Real y Lenguajes de Programación, Addison Wesley, Madrid.

Crespo, A., Alonso A. (2006). Una panorámica de los sistemas de tiempo real. Revista Iberoamericana de Automática e Informática Industrial, vol. 4, p. 8–18.

Frost, S. (2018). Introducción a la visión artificial. p. 1-24.

González, D., Cano, J., Guevara, P. (2017). Análisis comparativo de los tiempos de ejecución sobre SBC dos sistemas operativos de tiempo real. Pistas Educativas, vol. 39, 572-585.

Johansson, G. (2018). Real-Time Linux Testbench on Raspberry Pi 3 using Xenomai. Institute School of Electrical Engineering and Computer Science, Stockolm, p. 1-80.

Kanellakis, C., Nikolakopoulos, G. (2017). Survey on Computer Vision for UAVs: Current Developments and Trends. Journal of Intelligent & Robotic Systems, vol. 87, p. 141–168.

Sucar, E., Gómez, G. (2011). Visión computacional. Instituto Nacional de Astrofísica, Óptica y Electrónica, México.

Reghenzani, F., Massari, G., Fornaciari, W. (2019). The Real-Time Linux Kernel: A Survey on PREEMPT_RT. ACM Computing Surveys, vol. 52, p. 1-36.

Rodríguez, E., Bugarin, E., Rojas, J. A., Aguilar, A. (2021). Preliminary design and experimental tests of a real-time stereoscopic foveated vision system. Memorias del XXIII Congreso Mexicano de Robótica 2021, p. 26–31.

Ruijiang, L., Yan, G. (2001). Real-time stereo tracking of multiple moving heads. In Proceedings IEEE ICCV Workshop on Recognition, Analysis, and Tracking of Faces and Gestures in Real-Time Systems, p. 55-60.

Published
2024-04-22
How to Cite
Muñoz-Torres, J. L., Bugarin-Carlos, E., Rodríguez-Orozco, E., Rojas-Quintero, J. A., & Aguilar-Bustos, A. Y. (2024). Comparative of a real-time vision system under Xenomai and PREEMPT_RT. Pädi Boletín Científico De Ciencias Básicas E Ingenierías Del ICBI, 12(Especial2), 74-79. https://doi.org/10.29057/icbi.v12iEspecial2.12268