A georeferencing method for an open-pit mine surveying radar

Rossouw, Emile Francois (2011-03)

Thesis (MSc)--Stellenbosch University, 2011.

Thesis

ENGLISH ABSTRACT: Ground-based mobile mine survey radars are much more common now than they were a few years ago. Their ever growing popularity instigated the need for streamlining their operating procedures. One such a procedure is that of georeferencing the radar within the mine coordinate frame. Mine surveying radars have traditionally been georeferenced using a triangulation technique called survey resectioning, a time consuming process where both models are placed within the common coordinate frame by tying the models together with known targets or beacons. Survey resectioning requires surveying knowledge as well as access to a theodolite, an expensive high precision optical instrument used for measuring horizontal and vertical angles to the known targets. It is also sometimes necessary for this procedure to be performed in extreme weather conditions. Due to the limitations mentioned, this alignment method is not always practical or accurate if not performed correctly by the operator. In this thesis we investigate a new georeferencing scheme for ground-based mobile mine surveying radar, using a software-implementable three-dimensional model alignment. The scheme considers alignment complexity of four degrees of freedom and requires only an estimated radar position for complete convergence. The new scheme is tested on data previously georeferenced using the existing method.

AFRIKAANSE OPSOMMING: Die gewildheid van grondgebaseerde mobiele myn-radars is vinnig wˆereldwyd aan die toeneem. Hulle word veral in oopgroef-myne aangetref. Hierdie snelgroeiende gewildheid het dit genoodsaak om hul roetiene-gebruiksprosedures te vereenvoudig. ’n Voorbeeld van ’n vereenvoudigde prosedure is om die radar makliker binne die myn-ko¨ordinaatstelsel te posisioneer. Dit is van uiterste belang dat hierdie prosedure so vinnig en akkuraat as moontlik moet geskied. Voorheen is hierdie radar-eenhede deur middel van ’n omslagtige driehoeksmetingstegniek geposisioneer (dikwels in ongunstige weersomstandighede). Hierdie tegniek (genoem landmetingskorrelasie), is ’n proses waar die radar-eenheid in die myn se ko¨ordinaatstelsel geposisioneer word deur middel van bekende punte of bakens. Landmetingskorrelasie vereis boonop van die gebruiker om landmetingskennis te hˆe asook om duur en ingewikkelde toerusting soos ’n teodoliet te gebruik. ’n Teodoliet word algemeen gebruik vir die opmeting van horisontale- en vertikale hoeke na bekende punte. Bogenoemde beperkinge van landmetingskorrelasie het tot gevolg dat die moontlikheid van verkeerde posisionering bestaan indien die oprerateur onervare is. Die proses kan ook lank neem. In hierdie studie word ’n nuwe radarposisioneringsproses ondersoek wat deur middel van drie-dimensionele modelbelyning geimplementeer word as ’n rekenaarprogram. Die proses neem vier grade van vryheid in ag, en benodig slegs ’n benaderde posisie van die radar-eenheid vir konvergensie. Die nuwe proses is getoets op bestaande data wat deur middel van die landmetingskorrelasie-metode belyn was.

Please refer to this item in SUNScholar by using the following persistent URL: http://hdl.handle.net/10019.1/6907
This item appears in the following collections: