Search

Direktno georeferenciranje daljinsko vodenih letalnikov (angl. Unmanned Aerial Vehicle, UAV) se je v zadnjih letih izpopolnilo z vgrajenimi moduli izmere RTK (angl. Real-Time Kinematic) in PPK (angl. Post-Processed Kinematic), ki omogočajo določitev položaja v realnem času ali naknadno. Obe metodi imata svoje prednosti in slabosti, a se hkrati dopolnjujeta, ko ene izmed njiju zaradi različnih situacij (npr. lastnosti terena, ne/dostopnost mobilnega omrežja …) ni mogoče uporabiti. V raziskavi smo preizkušali letalnik Phantom 4 RTK na območju Strunjanskega klifa ob slovenski obali Jadranskega morja. Testno območje je zajemalo več snemalnih misij, katerih položaje smo določili v realnem času z navezavo na bazno postajo D-RTK 2 Mobile Station in naknadno na osnovi opazovanj GNSS z lastnih referenčnih baznih postaj v bližini snemanja. Položaje projekcijskih centrov posnetkov, pridobljene z metodama RTK in PPK, smo med sabo primerjali. Na podlagi koordinatnih razlik smo ugotavljali, ali različne lokacije baznih postaj vplivajo na kakovost rešitve PPK in ali se slednja značilno razlikuje od rešitve RTK. V članku je predstavljena tudi problematika obdelave PPK. Rezultati kažejo, da izbira lokacije bazne postaje lahko vpliva na kakovost določitve položaja UAV. Koordinatne razlike med rešitvami RTK in PPK v povprečju znašajo 0,3–8 cm v smereh E in ^ ter 1–6 cm v smeri ^. In recent years, direct georeferencing of Unmanned Aerial Vehicles (UAVs) has been perfected with built-in Real-Time Kinematic (RTK) and Post-Processed Kinematic (PPK) measurement modules, which enable real-time or subsequently positioning. Both methods have their advantages and disadvantages, but at the same time they complement each other when one of them cannot be used due to different situations (eg terrain characteristics, non / accessibility of the mobile network …). In the research, we tested the Phantom 4 RTK aircraft in the area of the Strunjan cliff along the Slovenian coast of the Adriatic Sea. The test area included several recording missions, the positions of which were determined in real time by connecting to the D-RTK 2 Mobile Station and subsequently based on GNSS observations from our own reference base stations in the vicinity of the survey. The positions of the projection centers of the images obtained by the RTK and PPK methods were compared with each other. Based on the coordinate differences, we determined whether different locations of base stations affect the quality of the PPK solution and whether the latter differs significantly from the RTK solution. The article also presents the issue of PPK processing. The results show that the choice of base station location can affect the quality of UAV positioning. The coordinate differences between the RTK and PPK solutions average 0.3–8 cm in the ^ and ^ directions and 1–6 cm in the ^ direction.

Ideja magistrske naloge je bila, da zgradimo delujoč nizkocenoven GNSS-sistem za nadgradnjo starejšega letalnika Phantom 4 Pro za namene direktnega georeferenciranja. V nalogi smo uporabili nizkocenoven sprejemnik GNSS, ZED-F9P, katerega delovanje smo ga testirali na površju in na letalniku. Testirali smo ga pri metodi izmere v realnem času (ang. real time kinematic – RTK) in metodi naknadne obdelave kinematičnih opazovanj GNSS (angl. post-processing kinematic – PPK). Pri poskusih na tleh smo koordinate zakoličenih točk najprej primerjali s koordinatami, pridobljenimi z nizkocenovnim sprejemnikom. Da bi ocenili točnost določitve koordinat točk trajektorije leta, smo letalnik opremili s 360° mini prizmo in izbranim točkam določili koordinate tudi s klasično terestrično izmero. Natančnost določitve koordinat na tleh in med letom v zraku je bila pri obeh kinematičnih metodah (RTK in PPK) manj kot 2 cm. Pri vseh štirih poskusih je točnost določitve koordinat za obe metodi PPK in RTK znašala manj kot 3 cm za horizontalni komponenti položaja in manj kot 6 cm za višino. Rezultati vzpostavljenega sistema kažejo, da lahko brezpilotne letalnike precej dobro nadgradimo s senzorji za pozicioniranje in spremljanje zasukov letalnika. V prihodnje bo potrebno razrešiti še problematiko časovne uskladitve fotografskih posnetkov in ostalih senzorjev na letalniku. In this master thesis, we pursued the idea of building a low-cost GNSS system and upgrading older Phantom 4 Pro aircraft for direct georeferencing purposes. In the thesis, a low-cost GNSS receiver ZED-F9 was tested both on the ground and on the aircraft. The receiver was tested using the real-time kinematic (RTK) and in post-processing kinematic (PPK) method. In the ground tests, the reference coordinates of the staked out points were compared first with the coordinates obtained using the low-cost GNSS receiver. To determine the flight trajectory position accuracy, a 360° mini prism was attached to the aircraft, which was measured using the classical terrestrial method. The accuracy of the determined coordinates, on the ground and on the aircraft, was less than 2 cm for both kinematic methods (RTK and PPK). In all four experiments, the position accuracy for the PPK and RTK methods was below 3 cm for horizontal components and less than 6 cm for the altitude. The results of the built low-cost GNSS system show that unmanned aerial vehicles can be quite well upgraded with sensors for positioning and monitoring aircraft rotations. In the future, it will be necessary to resolve the issue of time synchronisation of photographs and other sensors on the aircraft.