Mobile robotic platforms(MRP) are increasingly used in various areas of human activity. When using them, an important task is to determine the spatial orientation, measure the parameters of the movement of the MRP, etc. One of the problems that arises in determining navigation data and other measured parameters is their loss at a specific period, for example, due to interference or the temporary loss of visibility of GNSS navigation satellites. However, the functioning of navigation components of onboard radio-electronic devices for measuring movement parameters and determining the spatial orientation of the MRP requires the availability of primary navigation information without loss and in real time. Therefore, it is necessary to recover lost navigation data, especially in the case of MRPs, using onboard facilities with limited computing performance. Modern algorithms for recovering lost navigation data are analyzed, and it is determined that the publications do not pay enough attention to implementing these algorithms, taking into account the limitations of embedded systems. For implementation in mobile robotic platforms, an algorithm using the principal component analysis (PCA) method was selected, which, with low computational complexity, provides sufficient accuracy of data recovery. Using the developed algorithm in mobile robotic platforms provides data processing on a computing platform with limited resources. It allows streaming data to be processed on the MRP`s coordinates in real-time. Modern microcontrollers and systems on a chip (SoC) will enable you to solve the problem of recovering lost navigation data, taking into account restrictions on weight, dimensions, power consumption, etc. A structural diagram of a means for measuring motion parameters and determining spatial orientation for ground MRP has been developed. It is determined that the main components of the tool are a set of navigation sensors using a GPS/GNSS-based coordinate determination module. Data recovery tools have been created using the ESP32-C3 microcontroller, GNSS module type M10Q-5883, which contains a digital compass module QMC5883L and an accelerometer and gyroscope module MPU-6050. Debugging and testing the developed tools for recovering lost navigation data for MRPs have been performed. Analysis of the test results shows that the platform using the ESP32-C3 microcontroller provides data processing in 43 milliseconds. For the rate of GNSS data arrival at one measurement per second, this is enough to provide real-time mode.
[1] Tsmots I. G., Opotyak Yu. V., Shtogrinets B. V., Dzyuba A. O., Oliynyk Yu. Yu. (2023). Basic structure of the system of neurofuzzy control of a group of mobile robotic platforms. Ukrainian Journal of Information Technologies, Vol. 5, No. 1, 77–85. https://doi.org/10.23939/ujit2023.01.077
[2] Rejeb, A., Rejeb, K., Simske, S. J., & Treiblmaier, H. (2023). Drones for supply chain management and logistics: a review and research agenda. International Journal of Logistics Research and Applications, vol. 26, iss. 6, 708–731. DOI: 10.1080/13675567.2021.1981273
[3] Poulet, Guérin, F., and Guinand, F., (2021). Experimental and Simulation Platforms for Anonymous Robots Self- Localization, 29th Medit. Conf. on Control and Automation (MED), PUGLIA, Italy, 2021, 949–954. DOI: 10.1109/ MED51440.2021.9480244
[4] Izonin, I.; Kryvinska, N.; Tkachenko, R.; Zub, K. (2019). An approach towards missing data recovery within IoT smart system. Procedia Comput. Sci. J., 155, 11–18.
[5] Vedavalli, P., & Ch, D. (2023). A Deep Learning Based Data Recovery Approach for Missing and Erroneous Data of IoT Nodes. Sensors, 23(1), 170. https://doi.org/10.3390/ s23010170
[6] Gupta, G. P., Khandare, H. (2022). Missing Data Recovery Using Tensor Completion-Based Models for IoT-Based Air Quality Monitoring System. In: Karuppusamy, P., García Márquez, F. P., Nguyen, T. N. (eds) Ubiquitous Intelligent Systems. ICUIS 2021. Smart Innovation, Systems and Technologies, vol. 302. Springer, Singapore. https://doi.org/ 10.1007/978-981-19-2541-2_33
[7] Mondal, A., Das, M., Chatterjee, A., and Venkateswaran, P. (2020). Recovery of Missing Sensor Data by Reconstructing Time-varying Graph Signals, 30th European Signal Processing Conference (EUSIPCO), Belgrade, Serbia, pp. 2181–2185. DOI: 10.23919/EUSIPCO55093.2022.9909940
[8] Cheng, H., Wu, L., Li, R. et al. (2021). Data recovery in wireless sensor networks based on attribute correlation and extremely randomized trees. J Ambient Intell Human Comput 12, 245–259. DOI: https://doi.org/10.1007/s12652- 019-01475-z
[9] Ghorbani, B., Krishnan, S., and Zou, J. (2021). Debiasing stochastic gradient descent to handle missing values. Proc. 38th Int. Conf. Mach. Learn. (ICML), pp. 3748–3758.
[10] Tkachenko, R., Mishchuk, O., Izonin, I., Kryvinska, N., and Stoliarchuk, R. (2019). A non-iterative neural-like framework for missing data imputation. Procedia Computer Science, vol. 155, pp. 319–326.
[11] Tkachenko, R., Izonin, I. (2019). Model and principles for the implementation of neural-like structures based on geometric data transformations. Advances in Computer Science for Engineering and Education, Z. Hu, S. Petoukhov, I. Dychka, and M. He, Eds., Cham: Springer, 2019, pp. 578–587. DOI: 10.1007/978-3-319-91008-6_58.
[12] Lippi, V., Ceccarelli, G. (2019). Incremental principal component analysis: Exact implementation and continuity corrections. Proceedings of the 16th International Conference on Informatics in Control, Automation and Robotics (ICINCO 2019), Prague, Czech Republic, 2019, pp. 473–480. DOI: 10.5220/0007743604730480.
[13] Vaswani, N., Narayanamurthy, P. (2018). Static and dynamic robust PCA and matrix completion: A review. arXiv preprint arXiv:1803.00651, 2018 [Online]. https://arxiv.org/abs/1803. 00651.
[14] Ghahramani, Z., Hinton, G. E., (1996). Parameter estimation for linear dynamical systems. University of Toronto Technical Report CRG-TR-96-2, 1996.
[15] Tsmots, I., Skorokhoda, O., Tesliuk, T. & Rabyk, V. (2016). Designing features of hardware and software tools for intelligent processing of intensive data streams processing. IEEE First International Conference on Data Streams and Processing, DSMP, Lviv, pp. 332–335. DOI: 10.1109/ DSMP.2016.7583570