Professor Tanya Vladimirova’s research area is in custom computing for edge intelligence, which is intersection of embedded systems design, software, hardware, networking, digital signal processing, and artificial intelligence (AI).
Her expertise includes development of reliable intelligent embedded systems for AIoT (Artificial Intelligence of Things) that meet application requirements and covers the whole development cycle from algorithm investigation and modification through modelling to software implementation and hardware acceleration on embedded computing platforms such as Field Programmable Gate Arrays (FPGAs), Multi-core processors, and Graphical processing units (GPUs). There are numerous application areas for this type of systems including robotic systems, smart phones, aircraft, autonomous cars, and spacecraft; the Internet of things (IoT) being the largest and fast growing application field of embedded intelligence.
A key focus of Professor Vladimirova’s research has been the development of reliable and intelligent high-performance computing systems on board satellites, capable of withstanding the effects of some of the hazards of the space environment.
The research started at the Surrey Space Centre of the University of Surrey in Guildford, where Professor Vladimirova worked closely with the leading UK small satellite manufacturer, Surrey Satellite Technology Limited (SSTL). The research group that she founded and led there became internationally recognised for pioneering new ideas and innovative solutions, such as the following:
- the first satellite on-board computer implemented as a reconfigurable system-on-a-chip design;
- the first design and prototype of a femto-satellite based on a single printed circuit board (PCBSat);
- the first satellite-on-a-chip feasibility study;
- the first low power radiation hardening by design method using asynchronous logic;
- the first model of a space-based wireless sensor network using a shared distributed computing platform;
- the first fault-tolerant AES algorithm for encryption of satellite images on board;
- the first implementation of the CCSDS algorithm for lossless compression of multispectral satellite images on a reconfigurable FPGA; and
- the first feasibility study on flood monitoring using small satellites.
This work was continued at the School of Engineering of the University of Leicester, which Professor Vladimirova joined in 2011, where the following main achievements were accomplished:
- the first multi-sensor data fusion system in wireless sensor networks for interplanetary exploration;
- the first implementation of KLT-based adaptive lossless / lossy compression of hyperspectral satellite imagery on FPGAs;
- the first distributed Fault Detection Isolation and Recovery (FDIR) scheme for spaceborne reconfigurable multi FPGA-based systems ;
- the first design and prototype of a cooperative self-healing distributed intra-satellite computing system;
- the first distributed GNSS receiver for space-based wireless sensor networks; and
- the first design of an automated system for variable-rate fertilizer management using satellite imagery.
Current projects are focused on Machine Learning techniques for improving reliability of sensor data processing in Satellite Payloads and efficient image processing in Self-Driving Cars.
V. Afxentiou and T. Vladimirova, "Multi-Label Weather Image Classification for Embedded Computing Platforms”, Proc. of 20th IEEE Autonomous and Trusted Vehicles Conference (ATC 2023), 28-31 August 2023, Portsmouth, UK.
T. Turay and T. Vladimirova, " SSP Framework: A New Approach to Designing Lightweight Convolutional Neural Networks", IEEE Access, Vol. 10, 2022, pp. 102292 - 102307, doi: DOI: 10.1109/ACCESS.2022.3208924.
T. Turay and T. Vladimirova, "Towards Performing Image Classification and Object Detection with Convolutional Neural Networks in Autonomous Driving Systems: A Survey", IEEE Access, Vol. 10, 2022, pp. 14076-14119, doi: 10.1109/ACCESS.2022.3147495.
M. Fayyaz and T. Vladimirova “Survey and Future Directions of Fault-Tolerant Distributed Computing on Board Spacecraft”, Advances in Space Research, Vol. 58, Issue 11, 1 December 2016, pp. 2352–2375, Elsevier, doi: 10.1016/j.asr.2016.08.017.
F. Siegle, T. Vladimirova, J. Ilstad, and O. Emam. “Availability Analysis for Satellite Data Processing Systems Based on SRAM FPGAs”, IEEE Transactions on Aerospace and Electronic Systems, Vol. 52, Issue 3, 2016, pp. 977 - 989, doi: 10.1109/TAES.2016.140914.
X. Zhai and T. Vladimirova, “Efficient Data Processing Algorithms for Wireless Sensor Networks based Planetary Exploration”, AIAA Journal of Aerospace Information Systems, Vol. 13, No. 1, 2016, pp. 46-66.
F. Siegle, T. Vladimirova, O. Emam and J. Ilstad, "Mitigation of Radiation Effects in SRAM-based FPGAs for Space Applications", ACM Computing Surveys Journal, Vol. 47, issue 2, January 2015, article No. 37, pp. 37:1-37:34.
C.P.Bridges and T. Vladimirova. “Towards an Agent Computing Platform for Distributed Computing on Satellites”, IEEE Transactions on Aerospace and Electronic Systems, vol. 49, issue 3, pp. 1824-1838, July 2013.
S. Lopez, T. Vladimirova, C. González, J. Resano, D. Mozos and A. Plaza, “The Promise of Reconfigurable Computing for Hyperspectral Imaging On-Board Systems: Review and Trends”, IEEE Proceedings, special issue on "Information Processing of Very High Resolution Remote Sensing Data," Volume: 101 , Issue: 3, pp. 698 - 722, March 2013, IEEE.
N. R. Mat Noor and T. Vladimirova, “Investigation into Lossless Hyperspectral Image Compression for Satellite Remote Sensing”, International Journal of Remote Sensing, Volume 34, Issue 14, 2013, pp. 5072-5104, Taylor & Francis.
T. Vladimirova, M. Fayyaz, M.N.Sweeting, V.I. Vitanov, “A Novel Autonomous Low Cost On-Board Data Handling Architecture for a Pin-Point Planetary Lander”, Acta Astronautica, vol. 68, N 7-8, April-May 2011, pp. 811-829.
T. Vladimirova and S. Yuhaniz, “An Intelligent Decision-Making System for Flood Monitoring from Space”, Soft Computing Journal, Vol. 15, Issue 1, 2011, pp. 13-24, Springer.