Self-supervised monocular distance learning on a lightweight micro air vehicle

K. Lamers; Sjoerd Tijmons; Christophe de Wagter; Guido de Croon

doi:10.1109/IROS.2016.7759284

Self-supervised monocular distance learning on a lightweight micro air vehicle

K. Lamers, Sjoerd Tijmons, Christophe de Wagter, Guido de Croon

Control & Simulation

Research output: Chapter in Book/Conference proceedings/Edited volume › Conference contribution › Scientific › peer-review

10 Citations (Scopus)

56 Downloads (Pure)

Abstract

Obstacle detection by monocular vision is challenging because a single camera does not provide a direct measure for absolute distances to objects. A self-supervised learning approach is proposed that combines a camera and a very small short-range proximity sensor to find the relation between the appearance of objects in camera images and their corresponding distances. The method is efficient enough to run real time on a small camera system that can be carried onboard a lightweight MAV of 19 g. The effectiveness of the method is demonstrated by computer simulations and by experiments with the real platform in flight.

Original language	English
Title of host publication	IROS 2016
Subtitle of host publication	2016 IEEE/RSJ International Conference
Number of pages	6
DOIs	https://doi.org/10.1109/IROS.2016.7759284
Publication status	Published - 2016
Event	2016 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS 2016 - Daejeon, Korea, Republic of Duration: 9 Oct 2016 → 14 Oct 2016 http://www.iros2016.org/

Conference

Conference	2016 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS 2016
Abbreviated title	IROS 2016
Country/Territory	Korea, Republic of
City	Daejeon
Period	9/10/16 → 14/10/16
Internet address	http://www.iros2016.org/

Access to Document

10.1109/IROS.2016.7759284

IROS___Kevin___RevisionAccepted author manuscript, 2.22 MB

Cite this

@inproceedings{fd86be3eb80645acba12e447b255df00,

title = "Self-supervised monocular distance learning on a lightweight micro air vehicle",

abstract = "Obstacle detection by monocular vision is challenging because a single camera does not provide a direct measure for absolute distances to objects. A self-supervised learning approach is proposed that combines a camera and a very small short-range proximity sensor to find the relation between the appearance of objects in camera images and their corresponding distances. The method is efficient enough to run real time on a small camera system that can be carried onboard a lightweight MAV of 19 g. The effectiveness of the method is demonstrated by computer simulations and by experiments with the real platform in flight.",

author = "K. Lamers and Sjoerd Tijmons and {de Wagter}, Christophe and {de Croon}, Guido",

year = "2016",

doi = "10.1109/IROS.2016.7759284",

language = "English",

isbn = "978-1-5090-3762-9",

booktitle = "IROS 2016",

note = "2016 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS 2016, IROS 2016 ; Conference date: 09-10-2016 Through 14-10-2016",

url = "http://www.iros2016.org/",

}

TY - GEN

T1 - Self-supervised monocular distance learning on a lightweight micro air vehicle

AU - Lamers, K.

AU - Tijmons, Sjoerd

AU - de Wagter, Christophe

AU - de Croon, Guido

PY - 2016

Y1 - 2016

N2 - Obstacle detection by monocular vision is challenging because a single camera does not provide a direct measure for absolute distances to objects. A self-supervised learning approach is proposed that combines a camera and a very small short-range proximity sensor to find the relation between the appearance of objects in camera images and their corresponding distances. The method is efficient enough to run real time on a small camera system that can be carried onboard a lightweight MAV of 19 g. The effectiveness of the method is demonstrated by computer simulations and by experiments with the real platform in flight.

AB - Obstacle detection by monocular vision is challenging because a single camera does not provide a direct measure for absolute distances to objects. A self-supervised learning approach is proposed that combines a camera and a very small short-range proximity sensor to find the relation between the appearance of objects in camera images and their corresponding distances. The method is efficient enough to run real time on a small camera system that can be carried onboard a lightweight MAV of 19 g. The effectiveness of the method is demonstrated by computer simulations and by experiments with the real platform in flight.

U2 - 10.1109/IROS.2016.7759284

DO - 10.1109/IROS.2016.7759284

M3 - Conference contribution

SN - 978-1-5090-3762-9

BT - IROS 2016

T2 - 2016 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS 2016

Y2 - 9 October 2016 through 14 October 2016

ER -

Self-supervised monocular distance learning on a lightweight micro air vehicle

Abstract

Conference

Access to Document

Fingerprint

Cite this