Infrared horizon sensor for CubeSat implementation

dc.contributor.advisorSteyn, H. W.en_ZA
dc.contributor.authorWessels, Jurie Hendriken_ZA
dc.contributor.otherStellenbosch University. Faculty of Engineering. Dept. of Electrical and Electronic Engineering.en_ZA
dc.date.accessioned2018-02-20T09:35:09Z
dc.date.accessioned2018-04-09T07:01:04Z
dc.date.available2018-02-20T09:35:09Z
dc.date.available2018-04-09T07:01:04Z
dc.date.issued2018-03
dc.descriptionThesis (MEng)--Stellenbosch University, 2018.en_ZA
dc.description.abstractENGLISH ABSTRACT: Attitude determination is essential in satellite design, as it directly affects the pointing ability of the satellite. In the CubeSat industry there exists a need for high accuracy attitude sensors that are lowpower and low-cost. Currently, horizon sensors are a desirable option, but with recent growth in thermopile technology, it is possible to create horizon sensors that operate in the infrared spectrum offering tremendous benefits. This study focusses on the design, development, and evaluation of such an infrared horizon sensor. This includes the circuit and PCB design, software development and embedded implementation, as well as the creation of simulation/emulation environments for sensor calibration and evaluation. The critical limitation of this study is the extremely low resolution of the infrared camera (32 x 31 pixels) from which the horizon location should be determined to a sub-pixel accuracy. This limitation is overcome by calculating the gradient image with use of a Sobel Operator, after which the sub-pixel local extrema is determined by approximating a parabola shape on the horizon edge. In conclusion, a robust, low-power and low-cost sensor were developed, that is implementable on a Cube- Sat. This sensor delivers a worst case elevation accuracy of 0.075° with added noise of σ = 0:023°. Similarly, the rotation measurement delivers a worst case accuracy of 0.39° with added noise of σ = 0:14°. This satisfies the initial goal of reaching 0.1° elevation measurement accuracy.en_ZA
dc.description.abstractAFRIKAANSE OPSOMMING: Oriëntasiekennis is belangrik in satellietontwerp, aangesien dit sy oriëntasiebeheer direk beïnvloed. In die CubeSat-industrie bestaan daar 'n behoefte aan hoë akkuraatheid oriëntasiesensors wat lae krag en lae koste is. Tans is horison sensors 'n wenslike opsie, maar met 'n onlangse groei in termopiel tegnologie is dit moontlik om horison sensors te ontwerp wat in die infrarooi spektrum funksioneer, wat enorme voordele bied. Hierdie studie fokus op die ontwerp, ontwikkeling en evaluering van so 'n infrarooi horison sensor. Dit sluit in die voledige stroombaan ontwerp, sagteware-ontwikkeling en mikroverwerker implementering, asook die skep van simulasie- en emulasieomgewings vir sensorkalibrasie en evaluering. Die kritiese beperking van hierdie studie is die uiters lae resolusie van die infrarooi kamera (32x31 beeldpunte (Engels: pixels)) waarvan die horison-lokasie bepaal moet word vir 'n sub-beeldpunt-akkuraatheid. Hierdie beperking word oorkom deur die gradientbeeld te bereken deur gebruik te maak van 'n Sobel Operator, waarna die sub-beeldpunt plaaslike ekstrem bepaal word deur 'n paraboolvorm aan die horison rand te pas. Ten slotte is 'n robuuste, laekrag- en laekostesensor ontwikkel wat op 'n CubeSat implementeerbaar is. Hierdie sensor lewer 'n slegste geval elevasiemeting akkuraatheid van 0.075° met bygevoegde geraas van σ = 0:023°. Net so lewer die rotasiemeting 'n slegste geval akkuraatheid van 0.39° met bygevoegde geraas van σ = 0:14°. Dit voldoen aan die aanvanklike akkuraatheid doelwit om 'n 0.1° elevasiehoek te meet.af_ZA
dc.format.extent137 pages : illustrationsen_ZA
dc.identifier.urihttp://hdl.handle.net/10019.1/103564
dc.language.isoen_ZAen_ZA
dc.publisherStellenbosch : Stellenbosch Universityen_ZA
dc.rights.holderStellenbosch Universityen_ZA
dc.subjectCubeSaten_ZA
dc.subjectInfrared horizon sensorsen_ZA
dc.subjectUCTDen_ZA
dc.subjectSpectrum, Infrareden_ZA
dc.subjectUCTDen_ZA
dc.titleInfrared horizon sensor for CubeSat implementationen_ZA
dc.typeThesisen_ZA
Files
Original bundle
Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
wessels_infrared_2018.pdf
Size:
8.2 MB
Format:
Adobe Portable Document Format
Description:
License bundle
Now showing 1 - 1 of 1
Loading...
Thumbnail Image
Name:
license.txt
Size:
1.71 KB
Format:
Plain Text
Description: