SOFTWARE AND HARDWARE FOR MOBILE MEANS INFORMATION SUPPORT ONBOARD SYSTEMS WITH USE OF VISION SYSTEM

  • S.M. Sokolov Keldysh Institute of Applied Mathematics Russian Academy of Sciences
  • А. А. Boguslavsky Keldysh Institute of Applied Mathematics Russian Academy of Sciences
  • S.A. Romanenko Keldysh Institute of Applied Mathematics Russian Academy of Sciences
Keywords: Situational awareness, vision system, mobile robot, nonconventional calculators, algorithmic providing, airborne computers, real time

Abstract

Research objective is increase of autonomous mobile means information support efficiency with use of visual data and technologies of its development due to rational use of nonconventional calculators and special preparation of algorithmic providing. Use of nonconventional, heteroge-neous computing means allows to expand significantly a circle of visual data processing problems on the real time scale and, thereby, to increase situational awareness of autonomous robots and efficiency of its receiving. Rational use of nonconventional calculators demands essential altera-tion of algorithmic providing. The majority of visual data algorithms were developed counting on realization on traditional, von Neumann architecture of calculators and demand essential efforts for realization on parallel structures and developments of special programming tools. In work the emphasis on researches in the area use FPGA is placed and a number of approaches in special preparation of necessary algorithmic is considered. As model, demonstration examples of use of nonconventional calculators realization of such algorithms of visual data processing which areactively used in a wide range of information support problems of purposeful movements of mobile robot is considered. Preparation and realization on FPGA of such algorithms as creation of histo-grams, an optical flow calculation, segmentation of images is described. Results of experiments are given in the operating models of airborne computers. As basic data the visual data collected at the movement of mobile means in the conditions of habitat are used. All used software is executed on the basis of the unified software framework of the real time vision systems of domestic devel-opment. In the conclusion further steps in the specified direction, taking into account aspiration to use of domestic software and hardware are discussed.

References

1. Mar'yasin A.V., Manilenko I.N., Mikhaylov M.A., Perov I.A., Samarin O.F., Savost'yanov V.Yu., Il'in E.M. Tsifrovaya vychislitel'naya mashina dlya sovremennykh malogabaritnykh bortovykh radiolokatsionnykh stantsiy [Digital computer for modern small-size onboard radar stations], Vestnik SibGUTI [Vestnik SibGUTI], 2016, № 3, pp. 15-31.
2. Dashevskiy V.P., Bizin M.M. Obzor vozmozhnostey bortovykh vychisliteley [An overview of the capabilities of onboard computers], Doklady TUSURa [Proceedings of TUSUR], Sentyabr' 2015, No. 3 (37), pp. 924-929.
3. Lukin N.A. Bortovye funktsional'no-orientirovannye protsessory na osnove odnorodnykh vychislitel'nykh sred dlya mobil'nykh sistem real'nogo vremeni [Onboard functional-oriented processors based on homogeneous computing environments for mobile real-time systems], Fundamental'nye issledovaniya [Fundamental study], 2015, No. 12.
4. Chip Google Tensor Processor uprostit protsess mashinnogo obucheniya i vernet silu zakonu Mura. 3DNews [The Google Tensor Processor chip will simplify the machine learning process and restore the power of Moore's law. 3DNews]. Available at: https://3dnews.ru/933239.
5. Blok Google TPU vtorogo pokoleniya v zadachakh mashinnogo obucheniya demonstriruet proizvoditel'nost' vyshe, chem u GPU Nvidia GV100. iXBT.com [The second-generation Google TPU block in machine learning tasks demonstrates higher performance than the Nvidia GV100 GPU. iXBT.com]. Available at: https://www.ixbt.com/ news/2017/05/19/google-tpu-gpu-nvidia-gv100.html.
6. Podrobnosti o tenzornom soprotsessore Google TPU. Servernews [Details about the Google TPU tensor coprocessor. Servernews]. Available at: https://servernews.ru/957543.
7. Molaletsa Namoshe, Oduetse Matsebe and Nkgatho Tlale. Sensor Fusion and Its Applications, Feature extraction: techniques for landmark based navigation system, pp. 347-374.
8. Nützi G. et al. Fusion of IMU and Vision for Absolute Scale Estimation in Monocular SLAM, Journal of Intelligent & Robotic System. 2011. Vol. 61. pp. 287-299.
9. Engel J., Schöps T., Cremers D. LSD-SLAM: Large-Scale Direct Monocular SLAM, Proc. of the ECCV 2014, 2014, pp. 834-849.
10. Labbé M., Michaud F. RTAB-Map as an open-source lidar and visual simultaneous localiza-tion and mapping library, Journal of Field Robotics, 2018, pp. 1-31.
11. Gomez-Ojeda R. et al. PL-SLAM: A Stereo SLAM System Through the Combination of Points and Line Segments, IEEE Transactions on Robotics, 2019, Vol. 35, No. 3, pp. 734-746.
12. Galagan P., Kuz'minskiy L., Sorokin A. Reshenie zadach mashinnogo zreniya na baze geterogennoy platformy GRIFON [Solving machine vision problems based on the heterogene-ous GRYPHON platform], Control Engineering Rossiya [Control Engineering Russia], 2018, No. 1 (73), pp. 38-43.
13. Galagan P. Platforma GRIFON dlya resheniya zadach vstraivaemykh sistem spetsial'nogo naznacheniya [Gryphon platform for solving problems of embedded special purpose systems], Sovremennye tekhnologii avtomatizatsii [Modern automation technologies], 2015, No. 4.
14. Boguslavsky A.A., Sokolov S.M. Component Approach to the Applied Visual System Software Development, 7th World Multiconference on Systemics, Cybernetics and Informatics (SCI 2003), July 27-30, Orlando, Florida, USA, 2003.
15. Sokolov S.M., Boguslavsky A.A. Intellectual Images Processing for a Realtime Recognition Problem, Proc. The 2nd Intern. Multi-Conf. on Complexity, Informatics and Cybernetics (IMCIC2011), Orlando, Florida, USA, March 27th-30th, 2011, Orlando, Florida, USA, Vol. II, pp. 406-411.
16. Barlas G. Multicore and GPU Programming: An Integrated Approach. Morgan Kaufmann, 2014, 698 p.
17. Vasconcellos J.F. de A., Cáceres E.N., Mongelli H. and Song S.W. A Parallel Algorithm for Minimum Spanning Tree on GPU, 2017 International Symposium on Computer Architecture and High Performance Computing Workshops (SBAC-PADW), Campinas, 2017, pp. 67-72.
18. Vasconcellos J.F. de A., Cáceres E.N., Mongelli H., Song S.W., Dehne F., & Szwarcfiter J.L. New BSP/CGM algorithms for spanning trees, The International Journal of High Performance Computing Applications, 2019, Vol. 33 (3), pp. 444-461.
19. Pedro F. Felzenszwalb and Daniel P. Huttenlocher. Efficient Graph-Based Image Segmenta-tion, Int. J. Comput. Vision 59, 2 (September 2004), pp. 167-181.
20. Realizatsiya algoritma Kouskena [The implementation of the algorithm Koscina]. Available at: http://cs.brown.edu/people/pfelzens/segment/.
21. Borůvka Otakar. O jistém problému minimálním [About a certain minimal problem]. Práce Moravské přírodovědecké společnosti. V Brně III (in Czech and German), 1926, No. 3, pp. 37-58.
22. Proekty Gitlab: Realizatsiya algoritma Vasconcellos [GitLab projects: implementation of the Vasconcellos algorithm.]. Available at: https://gitlab.com/sergei-romanenko/vasconcellos-new-mst/ -/tree/master/img_seg_seq.
Published
2020-07-10
Issue
No 1 (2020)
Section
SECTION V. TECHNICAL VISION