PDF Upload: 



Akinyi, J, Mwaniki A, Gichamba A, Kariuki D, Chand P, Munene S, Nyakinyua C, Nzangi B, Akinyi V, Betsy M, Cosmas K, Mwangi M.  2021.  NanoSatellite Platform for the University of Nairobi (NaSPUoN) Student Project, 25-29 October 20. 72nd International Astronautical Congress (IAC). , Dubai, United Arab Emirates (presented online)


Cosmas, K, Kenichi A.  2020.  Utilization of FPGA for Onboard Inference of Landmark Localization in CNN-Based Spacecraft Pose Estimation. Aerospace. 7, Number 11 AbstractWebsite

In the recent past, research on the utilization of deep learning algorithms for space applications has been widespread. One of the areas where such algorithms are gaining attention is in spacecraft pose estimation, which is a fundamental requirement in many spacecraft rendezvous and navigation operations. Nevertheless, the application of such algorithms in space operations faces unique challenges compared to application in terrestrial operations. In the latter, they are facilitated by powerful computers, servers, and shared resources, such as cloud services. However, these resources are limited in space environment and spacecrafts. Hence, to take advantage of these algorithms, an on-board inferencing that is power- and cost-effective is required. This paper investigates the use of a hybrid Field Programmable Gate Array (FPGA) and Systems-on-Chip (SoC) device for efficient onboard inferencing of the Convolutional Neural Network (CNN) part of such pose estimation methods. In this study, Xilinx’s Zynq UltraScale+ MPSoC device is used and proposed as an effective onboard-inferencing solution. The performance of the onboard and computer inferencing is compared, and the effectiveness of the hybrid FPGA-CPU architecture is verified. The FPGA-based inference has comparable accuracy to the PC-based inference with an average RMS error difference of less than 0.55. Two CNN models that are based on encoder-decoder architecture have been investigated in this study and three approaches demonstrated for landmarks localization.


Fajardo, I, Lidtke AA, Bendoukha SA, Gonzalez-Llorente J, Rodríguez R, Morales R, Faizullin D, Matsuoka M, Urakami N, Kawauchi R, Miyazaki M, Yamagata N, Hatanaka K, Abdullah F, Rojas JJ, Keshk ME, Cosmas K, Ulambayar T, Saganti P, Holland D, Dachev T, Tuttle S, Dudziak R, Okuyama K-ichi.  2019.  Design, Implementation, and Operation of a Small Satellite Mission to Explore the Space Weather Effects in LEO. Aerospace. 6, Number 10 AbstractWebsite

Ten-Koh is a 23.5 kg, low-cost satellite developed to conduct space environment effects research in low-Earth orbit (LEO). Ten-Koh was developed primarily by students of the Kyushu Institute of Technology (Kyutech) and launched on 29 October 2018 on-board HII-A rocket F40, as a piggyback payload of JAXA’s Greenhouse gas Observing Satellite (GOSAT-2). The satellite carries a double Langmuir probe, CMOS-based particle detectors and a Liulin spectrometer as main payloads. This paper reviews the design of the mission, specifies the exact hardware used, and outlines the implementation and operation phases of the project. This work is intended as a reference that other aspiring satellite developers may use to increase their chances of success. Such a reference is expected to be particularly useful to other university teams, which will likely face the same challenges as the Ten-Koh team at Kyutech. Various on-orbit failures of the satellite are also discussed here in order to help avoid them in future small spacecraft. Applicability of small satellites to conduct space-weather research is also illustrated on the Ten-Koh example, which carried out simultaneous measurements with JAXA’s ARASE satellite.

Cosmas, K, Kenich A.  2019.  Implementation of Machine Learning Methods on FPGA for Onboard Satellite Operation, 21st – 25th Oct . 70th International Astronautical Congress (IAC). , Washington DC, USA


Kiruki, C.  2018.  Arid and Semi-Arid Lands Satellite (ASAL-SAT): A LoRa ground sensor network for easing life in Sub-Saharan Africa ASAL areas, 19th Nov 2018. 6th UNISEC-Global Meeting. , Strasbourg, France
Cosmas, K, Asami K.  2018.  Flexible Modularized Artificial Neural Network Implementation on FPGA. 2018 5th International Conference on Soft Computing Machine Intelligence (ISCMI). :1-5. Abstract


UoN Websites Search