Camera-based Obstacle Classification for Automated Reach Trucks Using Deep Learning

Marian Himstedt; Erik Maehle

Camera-based Obstacle Classification for Automated Reach Trucks Using Deep Learning

Marian Himstedt, Erik Maehle

Institute of Computer Engineering

Institute of Computer Engineering

Abstract

This paper focuses on the classification of obstacles that are widely present in warehouse environments using an RGBD camera. Our approach applies depth segmentation to detect obstacles which are classified using a Convolutional Neural Network and a Support Vector Machine. Our system is evaluated on real-world data captured from an automated reach truck in a warehouse environment.

Original language	English
Title of host publication	Proceedings of ISR 2016: 47st International Symposium on Robotics
Editors	Marian Himstedt, Erik Maehle
Publisher	IEEE
Publication date	01.01.2016
ISBN (Print)	978-3-8007-4231-8
Publication status	Published - 01.01.2016
Event	Proceedings of ISR 2016: 47st International Symposium on Robotics - München, Germany Duration: 21.06.2016 → 22.06.2016 https://ieeexplore.ieee.org/xpl/mostRecentIssue.jsp?punumber=7558447

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

https://ieeexplore.ieee.org/xpl/mostRecentIssue.jsp?punumber=7558447

Cite this

@inproceedings{8e3e90f277ae4b679bf5859cb54418c3,

title = "Camera-based Obstacle Classification for Automated Reach Trucks Using Deep Learning",

abstract = "This paper focuses on the classification of obstacles that are widely present in warehouse environments using an RGBD camera. Our approach applies depth segmentation to detect obstacles which are classified using a Convolutional Neural Network and a Support Vector Machine. Our system is evaluated on real-world data captured from an automated reach truck in a warehouse environment.",

author = "Marian Himstedt and Erik Maehle",

year = "2016",

month = jan,

day = "1",

language = "English",

isbn = "978-3-8007-4231-8",

editor = "Marian Himstedt and Erik Maehle",

booktitle = "Proceedings of ISR 2016: 47st International Symposium on Robotics",

publisher = "IEEE",

note = "Proceedings of ISR 2016: 47st International Symposium on Robotics ; Conference date: 21-06-2016 Through 22-06-2016",

url = "https://ieeexplore.ieee.org/xpl/mostRecentIssue.jsp?punumber=7558447",

}

Himstedt, M & Maehle, E 2016, Camera-based Obstacle Classification for Automated Reach Trucks Using Deep Learning. in M Himstedt & E Maehle (eds), Proceedings of ISR 2016: 47st International Symposium on Robotics. IEEE, Proceedings of ISR 2016: 47st International Symposium on Robotics, München, Germany, 21.06.16. <https://ieeexplore.ieee.org/xpl/mostRecentIssue.jsp?punumber=7558447>

TY - GEN

T1 - Camera-based Obstacle Classification for Automated Reach Trucks Using Deep Learning

AU - Himstedt, Marian

AU - Maehle, Erik

PY - 2016/1/1

Y1 - 2016/1/1

N2 - This paper focuses on the classification of obstacles that are widely present in warehouse environments using an RGBD camera. Our approach applies depth segmentation to detect obstacles which are classified using a Convolutional Neural Network and a Support Vector Machine. Our system is evaluated on real-world data captured from an automated reach truck in a warehouse environment.

AB - This paper focuses on the classification of obstacles that are widely present in warehouse environments using an RGBD camera. Our approach applies depth segmentation to detect obstacles which are classified using a Convolutional Neural Network and a Support Vector Machine. Our system is evaluated on real-world data captured from an automated reach truck in a warehouse environment.

M3 - Conference contribution

SN - 978-3-8007-4231-8

BT - Proceedings of ISR 2016: 47st International Symposium on Robotics

A2 - Himstedt, Marian

A2 - Maehle, Erik

PB - IEEE

T2 - Proceedings of ISR 2016: 47st International Symposium on Robotics

Y2 - 21 June 2016 through 22 June 2016

ER -

Camera-based Obstacle Classification for Automated Reach Trucks Using Deep Learning

Abstract

UN SDGs

Access to Document

Fingerprint

Cite this