ICONS 2019 Paper Abstract


Paper WeA1Rob.2

Hiba, Antal (Hungarian Academy of Sciences, Institute for Computer Science an), Aleksziev, Rita (Institute for Computer Science and Control, Hungarian Academy of), Pázmán, Koppány (Institute for Computer Science and Control, Hungarian Academy of), Bauer, Peter (Institute for Computer Science and Control,HungarianAcademyof Sc), Benczúr, András (Institute for Computer Science and Control, Hungarian Academy of), Zarandy, Akos (MTA-SZTAKI), Daróczy, Bálint (Institute for Computer Science and Control, Hungarian Academy of)

The Applicability of On-Line Contextual Calibration to a Neural Network Based Monocular Collision Avoidance System on a UAV

Scheduled for presentation during the Regular Session "Robotics and Autonomous Vehicles" (WeA1Rob), Wednesday, August 21, 2019, 11:10−11:30,

5th IFAC International Conference on Intelligent Control and Automation Sciences, August 21-23, 2019, Queen’s University Belfast, Northern Ireland

This information is tentative and subject to change. Compiled on October 16, 2021

Keywords Learning, adaptation and evaluation, Aerospace, Robotics and autonomous systems


Contextual calibration for object detection is a technique where a pretrained network collects attractive false positives during a calibration phase and use this calibration data for further training. This paper investigates the applicability of this method to a vision based on-board sense and avoid system, which requires intruder aircraft detection in camera images. Various landscape and sky backgrounds were generated by Unreal4 3D engine for calibration tests. Contextual calibration is a promising candidate for handling extreme situations which are not covered well in the training data.


Technical Content Copyright © IFAC. All rights reserved.

This site is protected by copyright and trademark laws under US and International law.
All rights reserved. © 2002-2021 PaperCept, Inc.
Page generated 2021-10-16  23:09:02 PST   Terms of use