MODELS AND METHODS OF INTELLECTUAL INFORMATION TECHNOLOGY OF AUTONOMOUS NAVIGATION FOR COMPACT DRONES

V. V. Moskalenko, A. S. Moskalenko, A. G. Korobov

Abstract


Context of this article topics is that the issues of choosing the optimal in information and cost sense of models and methods of data
analysis in autonomous navigation systems under a priori uncertainty, resource and information constraints are not sufficiently investigated
and are still not fully resolved.
Objective – to increase the efficiency of the autonomous navigation system of a compact drone on the terrain, based on data from visual
and inertial sensors, operating in training and exam modes in the conditions of limited computing resources and the volume of the training sample, in the information and value sense.
Methods of a research are based on the usage of technology of convolutional neural networks for the formation of a feature representation of visual observations, sparse-coding neural gas algorithms for the training of convolutional filters, the model of support vectors for regression analysis of data, on the principles of mathematical statistics and information theory for constructing and evaluating the functional efficiency of classification decision rules.
Results: new models and methods of information intelligent technology of autonomous navigation for compact drones have been developed,
allowing the training of the most computationally labor-intensive component of the system, the feature extractor from observation, in
unsupervised manner in the process of direct propagation of the signal. In this case, the criterion for choosing the optimal parameters in information and cost sense for model of data analysis is proposed, and it is shown by the results of physical modeling that the validity of the
formed decision rules is acceptable for practical use.
Conclusions. The architecture of the convolutional neural network and the method of its unsupervised learning for the formation of a
feature representation of observations in the autonomous navigation problem based on the algorithm of sparse coding neural gas is proposed. The criterion for choosing data analysis parameters is developed and according to the results of physical modeling, the suitability for practical use of developed algorithms of navigation on unknown terrain is proved. The practical value of the results obtained for unmanned aviation is to form a modern scientific and methodological basis for designing capable of training autonomous navigation systems for compact drones operating in the conditions of resource and information constraints.

Keywords


navigation; visual odometry; unmanned aerial vehicle; convolutional neural network; neural gas; information criterion; support vector method.

References


Wang S., Deng Z. , Yin G. An Accurate GPS-IMU/DR Data Fusion Method for Driverless Car Based on a Set of Predictive Models and Grid Constraints. Sensors [Electronic resource]. Basel, Switzerland, Multidisciplinary Digital Publishing Institute, 2016, Vol. 16(3), pp. 280–293. DOI 10.3390/s16030280.

Suwandi B., Kitasuka T., Aritsugi M. Low-cost IMU and GPS

fusion strategy for apron vehicle positioning, TENCON 2017 IEEE Region 10 Conference, 2017, Penang, Malaysia, 5–8 Nov. 2017 : proceedings, pp. 449–454. DOI: 10.1109/TENCON.2017.8227906.

Mary B., Gerhard P. H. A. Pose Estimation of a Mobile Robot

Based on Fusion of IMU Data and Vision Data Using an Extended Kalman Filter. [Electronic resource], Sensors, 2017, Vol. 17(10), P. 2164. DOI: 10.3390/s17102164.

Folkesson J. , Leederkerken J. , Williams R. , Patrikalakis A. A Feature Based Navigation System for an Autonomous Underwater Robot, In: Laugier C., Siegwart R. (eds) Field and Service Robotics. Springer Tracts in Advanced Robotics. Springer, Berlin, Heidelberg, 2008, Vol. 42, pp. 105–114. DOI 10.1007/978-3-540-75404-6_10.

Fanani N., Stürck A., Ochs M., Bradler H. , Mester R. Predictive monocular odometry (PMO): What is possible without RANSAC and multiframe bundle adjustment? Image and Vision Computing volume, 2017, Vol. 68, pp. 3–13. DOI:

1016/j.imavis.2017.08.002. 6. Cadena C., Carlone L., Carrillo H. et al Past, Present, and Future of Simultaneous Localization And Mapping: Towards the Robust-Perception Age, IEEE Transactions on Robotics, 2016, Vol 32, No. 6, pp. 1309–1332. DOI: 10.1109/ TRO.2016.2624754.

Dorian G.-L., Marta S. , Juan D. T., Montiel J.M.M. Real-time

monocular object SLAM, Robotics and Autonomous Systems,

, Vol. 75, І. PB, pp. 435–449. DOI:10.1016/

j.robot.2015.08.009.

Ayoul Th., Buckley T., Crevier F. UAV Navigation above Roads Using Convolutional Neural Networks [Electronic resource], Available from: http://cs231n.stanford.edu/reports/

/pdfs/553.pdf.

Mohanty V. DeepVO : A Deep Learning approach for Monocular Visual Odometry [Electronic resource], Computer Vision and Pattern Recognition, 2016, Available from: https://arxiv.org/pdf/1611.06069.pdf.

Labusch K., Barth E., Martinetz T. Sparse Coding Neural Gas: Learning of Overcomplete Data Representations, Neurocomputting, 2009, Vol. 72, І. 7–9, pp. 1547–1555.

DOI:10.1016/j.neucom.2008.11.027.

Xu H., R. Wang, Wang K. A New SVR Incremental Algorithm

Based on Boundary Vector [Electronic resource], International

Conference on Computational Intelligence and Software Engineering (CiSE 2010), Wuhan, China 10–12 Dec. 2010, proceedings. Zürich, Springer, 2010, pp. 145–149.

DOI: 10.1109/CISE.2010.5676955.

Moskalenko V. V., Korobov A. G. Information-extreme algorithm of the system for recognition of objects on the terrain with optimization parameter feature extraction, Radio Electronics, Computer Science, Control, 2017, No. 2, pp. 38–45. DOI 10.15588/1607-3274-2017-2-7.

Moskalenko V. V., Rizhova A. S., Dovbysh A. S. Intelektual’na systema pidtrymky pryjnjattja rishen’ dlja funkcional’nogo diagnostuvannja na gamma-kameri, Radio Electronics, Computer Science, Control, 2015, No. 4, pp. 52–58. DOI: 10.15588/1607-3274-2015-4-8.


GOST Style Citations


1. Wang S. An Accurate GPS-IMU/DR Data Fusion Method for
Driverless Car Based on a Set of Predictive Models and Grid
Constraints. Sensors [Electronic resource] / S. Wang, Z. Deng,
G. Yin // Basel, Switzerland : Multidisciplinary Digital Publishing
Institute. – 2016. – Vol. 16(3). – P. 280–293.
DOI 10.3390/s16030280.
2. Suwandi B. Low-cost IMU and GPS fusion strategy for apron
vehicle positioning. / B. Suwandi, T. Kitasuka, M. Aritsugi //
TENCON 2017 IEEE Region 10 Conference, 2017, Penang,
Malaysia, 5–8 Nov. 2017 : proceedings. – P. 449–454.
DOI: 10.1109/TENCON.2017.8227906.
3. Mary B. A. Pose Estimation of a Mobile Robot Based on Fusion
of IMU Data and Vision Data Using an Extended Kalman Filter.
[Electronic resource] / B. A. Mary, P. H. Gerhard //
Sensors. – 2017. – Vol. 17(10). – P. 2164. DOI:
10.3390/s17102164.
4. A Feature Based Navigation System for an Autonomous Underwater
Robot / J. Folkesson, J. Leederkerken, R. Williams,
A. Patrikalakis // In: Laugier C., Siegwart R. (eds) Field and
Service Robotics. Springer Tracts in Advanced Robotics. –
Springer : Berlin, Heidelberg. – 2008. – Vol. 42. – P. 105–114.
DOI 10.1007/978-3-540-75404-6_10.
5. Predictive monocular odometry (PMO): What is possible without
RANSAC and multiframe bundle adjustment? / N. Fanani,
A. Stürck, M. Ochs, H. Bradler, R. Mester // Image and Vision
Computing volume. – 2017. – Vol. 68. – P. 3–13 DOI:
10.1016/j.imavis.2017.08.002.
6. Past, Present, and Future of Simultaneous Localization And
Mapping: Towards the Robust-Perception Age / C. Cadena, L.
Carlone, H. Carrillo et al // IEEE Transactions on Robotics. –
2016. – Vol 32. – No. 6. – P. 1309–1332.
DOI: 10.1109/TRO.2016.2624754.
7. Dorian G.-L. Real-time monocular object SLAM /
G.-L. Dorian, S. Marta, D. T. Juan, J.M.M. Montiel // Robotics
and Autonomous Systems. – 2016. – Vol. 75. – І. PB. –
P. 435–449. DOI: 10.1016/j.robot.2015.08.009.
8. Ayoul Th. UAV Navigation above Roads Using Convolutional
Neural Networks [Electronic resource] / Th. Ayoul, T. Buckley,
F. Crevier // Available from: http://cs231n.stanford.edu/reports/
2017/pdfs/553.pdf.
9. Mohanty V. DeepVO : A Deep Learning approach for Monocular
Visual Odometry [Electronic resource] / V. Mohanty //
Computer Vision and Pattern Recognition. – 2016. Available
from: https://arxiv.org/pdf/ 1611.06069.pdf.
10. Labusch K. Sparse Coding Neural Gas: Learning of Overcomplete
Data Representations / K. Labusch, E. Barth, T. Martinetz
// Neurocomputing. – 2009. – Vol. 72. – І. 7–9. – P. 1547–1555.
DOI:10.1016/j.neucom.2008.11.027.
11. Xu H. A New SVR Incremental Algorithm Based on Boundary
Vector [Electronic resource] / H. Xu, R. Wang, K. Wang // International
Conference on Computational Intelligence and
Software Engineering (CiSE 2010), Wuhan, China 10 12 Dec.
2010 : proceedings. – Zürich : Springer, 2010. – P. 145 – 149.
DOI: 10.1109/CISE.2010.5676955.
12. Moskalenko V. V. Information-extreme algorithm of the system
for recognition of objects on the terrain with optimization parameter
feature extraction. / V. V. Moskalenko, A. G. Korobov.
// Radio Electronics, Computer Science, Control. – 2017. –
No 2. – P. 38–45. DOI 10.15588/1607-3274-2017-2-7.
13. Москаленко В. В. Інтелектуальна система підтримки
прийняття рішень для функціонального діагностування на
гамма-камері / В. В. Москаленко, А. С. Рижова, А. С. Довбиш
// Радіоелектроніка, інформатика, управління. – 2015. – № 4. –
С. 52–58. DOI: 10.15588/1607-3274-2015-4-8.






Copyright (c) 2018 V. V. Moskalenko, A. S. Moskalenko, A. G. Korobov

Creative Commons License
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.

Address of the journal editorial office:
Editorial office of the journal «Radio Electronics, Computer Science, Control»,
Zaporizhzhya National Technical University, 
Zhukovskiy street, 64, Zaporizhzhya, 69063, Ukraine. 
Telephone: +38-061-769-82-96 – the Editing and Publishing Department.
E-mail: rvv@zntu.edu.ua

The reference to the journal is obligatory in the cases of complete or partial use of its materials.