Detekcija ovir z zlivanjem senzorske informacije za avtonomna plovila : diplomsko delo

OVEN, FRANC (author), Kristan, Matej (mentor)

V diplomskem delu obravnavamo problem avtonomne navigacije plovil. Predpostavljamo, da je plovilo opremljeno s senzorji kot so GPS, IMU, kompas, stereo sistem dveh kamer ter procesno enoto na kateri se izvajajo algoritmi segmentacije in planiranja poti. Začnemo s predstavitvijo algoritma segmentacije, ki skupaj z naprednimi metodami računalniškega vida iz slik kamer oceni morebitne ovire (boje, plavalci, ladje, obala). Osredotočimo se na del, kjer s pomočjo meritev inercijske merilne naprave ocenimo horizont na slikah kamer in s tem pripomoremo k robustnejši oceni morja. Za izboljšano lokalizacijo ovir, s pomočjo kalibriranega stereo sistema kamer ocenimo oddaljenost pikslov na sliki v 3D prostoru. Slike morajo biti za računanje globine rekticirane, saj s tem poenostavimo iskanje korespondenc na enodimenzionalen problem. V nadaljevanju predstavimo, implementiramo in ovrednotimo nase predloge za izboljsavo ter predstavimo smernice za nadaljni razvoj.

• Language: Slovenian
• Keywords: stereo sistem, 3D inferenca, inercijska merilna naprava, hori- zont, segmentacija, računalništvo, računalništvo in informatika, univerzitetni študij, diplomske naloge
• Work type: Bachelor thesis/paper
• Typology: 2.12 - Final Research Report
• Organization: FRI - Faculty of Computer and Information Science
• Publisher: [F. Oven]
• Year: 2014
• Number of pages: 58 str.
• PID: 20.500.12556/RUL-29562
• COBISS.SI-ID: 1536127939
• Publication date in RUL: 23.09.2014

OVEN, FRANC, 2014, Detekcija ovir z zlivanjem senzorske informacije za avtonomna plovila : diplomsko delo [online]. Final Research Report. F. Oven. [Accessed 25 March 2025]. Retrieved from: https://repozitorij.uni-lj.si/IzpisGradiva.php?lang=eng&id=29562

Secondary language

• Language: English
• Title: Obstacle detection through sensor fusion is unmanned surface vehicles

Abstract

In our diploma thesis we issue a navigation problem of autonomous surface vehicles. Presume we have a vessel appointed with sensors, such as GPS, IMU, compass, a stereo sistem of two cameras and a standalone processing unit that combines sensor data and performs segmentation and planning in real time. We begin by introducing the algorithm of segmentation, that along with help of advanced computer vision methods extracts useful visual information, therefore avoids obstacles in its path (boats, swimmers, buoys). Then, we focus on horizon estimation by taking into account data of innertial measurement unit, with whom we improve the estimation of a sea border. To improve localization of obstacles in front of the vessel, we calculate depth of every pixel in the image. Image pairs are first rectified as we simplify the correspondence search to single dimension. Furthermore, we present, implement and evaluate our methods. We conclude by discussing further work that can be done.

• Keywords: stereo system, 3D inference, inertial measurement unit, horizon, seg- mentation, computer science, computer and information science, diploma