Implementation of the visual data processing algorithms for onboard computing units

Sergey M. Sokolov
Doctor of Physical and Mathematical Sciences, Professor, Keldysh Institute of Applied Mathematics, Chief Research Scientist, 4, Miusskaya pl., Moscow, 125047, Russia, tel.: +7(916)122-01-13, This email address is being protected from spambots. You need JavaScript enabled to view it., ORCID: 0000-0001-6923-2510

Andrey A. Boguslavsky
Doctor of Physical and Mathematical Sciences, Keldysh Institute of Applied Mathematics, Leading Research Scientist, 4, Miusskaya pl., Moscow, 125047, Russia, tel.: +7(499)250-78-73, This email address is being protected from spambots. You need JavaScript enabled to view it., ORCID: 0000-0001-7560-339X

Sergei A. Romanenko
PhD in Physics and Mathematics, Keldysh Institute of Applied Mathematics, Leading Research Scientist, 4, Miusskaya pl., Moscow, 125047, Russia, tel.: +7(499)250-78-73, This email address is being protected from spambots. You need JavaScript enabled to view it.

Received 15 October 2020

Abstract
According to the short analysis of modern experience of hardware and software for autonomous mobile robots a role of computer vision systems in the structure of those robots is considered. A number of configurations of onboard computers and implementation of algorithms for visual data capturing and processing are described. In original configuration space the «algorithms-hardware» plane is considered. For software designing the realtime vision system framework is used. Experiments with the computing module based on the Intel/Altera Cyclone IV FPGA (implementation of the histogram computation algorithm and the Canny's algorithm), with the computing module based on the Xilinx FPGA (implementation of a sparse and dense optical flow algorithms) are described. Also implementation of algorithm of graph segmentation of grayscale images is considered and analyzed. Results of the first experiments are presented.

Key words
Mobile robots, high autonomy degree, information support of robots, computer vision systems, intellectual algorithms, high-performance computations.

Acknowledgements
This work was supported by the RFBR grant no. 19-07-01113.

DOI
https://doi.org/10.31776/RTCJ.9204

Bibliographic description
Sokolov, S., Boguslavsky, A. and Romanenko, S., 2021. Implementation of the visual data processing algorithms for onboard computing units. Robotics and Technical Cybernetics, 9(2), pp.106-111.

UDC identifier:
004.4:004.93'1

References

1. Braden, B., 2003. A view that is out of this world. Univ. of Regina, Campion’s Brag, 9, pp.6-8.
2. Lakdawalla, E., 2018. The Design and Engineering of Curiosity How the Mars Rover Performs Its Job. Wiley Publ.
3. NVIDIA. n.d. Self-Driving Cars Technology & Solutions from NVIDIA Automotive. [online] Available at: <https://www.nvidia.com/en-us/self-driving-cars/drive-platform> [Accessed 10 October 2020].
4. Intel Corp. 2020. Intel GO Automated Driving Solutions. [online] Available at: <https://newsroom.intel.com/newsroom/wp-content/uploads/sites/11/2017/01/intel-go-product-brief.pdf> [Accessed 10 October 2020].
5. En.wikichip.org. 2020. FSD Chip - Tesla - WikiChip. [online] Available at: <https://en.wikichip.org/wiki/tesla_(car_company)/fsd_chip> [Accessed 10 October 2020].
6. Neosys Tech. 2020. Edge AI GPU Computing. [online] Available at: <https://www.neousys-tech.com/en/product/application/edge-ai-gpu-computing> [Accessed 10 October 2020].
7. Apollo. 2020. Baidu, Apollo Open Platform. [online] Available at: <http://apollo.auto/platform/hardware.html> [Accessed 10 October 2020].
8. Galagan, P., Kuz'minskiy, L. and Sorokin, A., 2018. Reshenie zadach mashinnogo zreniya na baze geterogennoy platformy GRIFON [Solving machine vision problems based on the heterogeneous platform GRIFON]. Control Engineering, 1(73), pp. 38-43. (in Russian).
9. Boguslavskiy, A.A., Borovin, G.K., Kartashev, V.A., Pavlovskiy, V.E. and Sokolov, S.M., 2019. Modeli i Algoritmy dlya Intellektual'nykh Sistem Upravleniya [Models and Algorithms for Intelligent Control Systems]. Moscow: IAM n.a. Keldysh Publ., p.235. (in Russian).
10. Sokolov, S.M. and Boguslavsky, A.A., 2016. Methodological aspects for the development of information systems of unmanned mobile vehicles. In: Proc. of the 13th International Conference on Informatics in Control, Automation and Robotics (ICINCO2016), vol.2, pp.492-498.
11. Sokolov, S.M., Boguslavsky, A.A., Vasilyev, A.I. and Trifonov, O.V., 2013. Development of software and hardware of entry-level vision systems for navigation tasks and measuring. Advances in Intelligent Systems and Computing (Springer), 208, pp.463-476.
12. Felzenszwalb, P. and Huttenlocher, D., 2004. Efficient Graph-Based Image Segmentation. Int. J. Comput. Vision, 59(2), pp.167-181.
13. Vasconcellos, J., Caceres, Е., Mongelli, Н. and Song, S., 2017. А Parallel Algorithm for Minimum Spanning Tree on GPU. In: Proc of the lnternational Symposium оп Computer Architecture and High Performance Computing Workshops (SBAC-PADW), рр. 67-72.