izpis_h1_title_alt

Detekcija in sledenje oznake za avtonomno pristajanje brezpilotnega letalnika
ID Volk, Matic (Author), ID Skočaj, Danijel (Mentor) More about this mentor... This link opens in a new window

.pdfPDF - Presentation file, Download (10,34 MB)
MD5: AB945B02372C7ECC924CD7DA64118F08
PID: 20.500.12556/rul/66ef0ae4-f272-428c-aaa9-e4ffeee0a837

Abstract
Brezpilotni letalniki so v zadnjih letih postali pomembno sredstvo za uporabo tako v vojaške kot tudi v civilne namene. V magistrski nalogi razvijemo sistem za detekcijo in sledenje oznake, ki bo pripomogel k realizaciji avtonomnega pristajanja brezpilotnega letalnika. Trenutni letalnik je omejen na pristajanje s padalom, kar onemogoča pristajanje na premikajočih platformah in lokacijah z majhno pristajalno površino. Za odpravo teh težav smo si zastavili cilj razviti sistem, ki bi letalniku omogočili pristajanje v mrežo. Mreža je označena z zasnovano oznako. Razviti sistem smo ovrednotili na posebej pripravljenih videoposnetkih, ki simulirajo realno okolje. Rezultati so pokazali, da je delovanje sistema zadovoljivo pod različnimi pogoji, kot so različni koti, razdalje, osvetljenosti, zamegljenosti, okolja in dodajanje megle v pripravljenih videoposnetkih.

Language:Slovenian
Keywords:brezpilotni letalnik, računalniški vid, oznaka, detekcija, sledenje
Work type:Master's thesis/paper
Organization:FRI - Faculty of Computer and Information Science
Year:2017
PID:20.500.12556/RUL-99050 This link opens in a new window
Publication date in RUL:22.12.2017
Views:1549
Downloads:640
Metadata:XML DC-XML DC-RDF
:
Copy citation
Share:Bookmark and Share

Secondary language

Language:English
Title:Marker detection and tracking for autonomous landing of unmanned aerial vehicle
Abstract:
In recent years the unmanned aerial vehicles (UAVs) became important assets for military and civil use. In our master thesis we developed a system for marker detection and tracking that enables landing of a UAV without a parachute. The current UAV is limited to landing with a parachute, which makes it difficult to land on moving platforms and small areas. The goal was to develop a system that can help the UAV to land into the net markerd with a purposely designed marker. The developed system was evaluated on captured videos that simulate the real environment. Results have shown that the system works well under different circumstances such as various angles, distances, illuminations, blurriness, surrounding areas and fogginess.

Keywords:unmanned aerial vehicle, computer vision, marker, detection, tracking

Similar documents

Similar works from RUL:
Similar works from other Slovenian collections:

Back