DocumentCode :
1941642
Title :
A Smart vision system for advanced LGV navigation and obstacle detection
Author :
Massimo, Bertozzi ; Luca, Bombini ; Alberto, Broggi ; Alessandro, Coati
Author_Institution :
Dept. of Inf. Eng., Univ. of Parma, Parma, Italy
fYear :
2012
fDate :
16-19 Sept. 2012
Firstpage :
508
Lastpage :
513
Abstract :
This article presents the VisLab solution for obstacle detection and navigation support for unmanned vehicles in industrial environments. Although the literature contains many examples to tackle this problem, this solution can be considered innovative as it improves traditional laser-based systems. The proposed system is composed by two sub-systems. The first one is an obstacle detection system, which also allows the detection of hanging obstacles, within a 3D monitored area. This solution outperforms the original laser scanner based system used for safety which was limited to bi-dimensional areas only. Another vision system is used for tracking a guideline on the ground, that solves problems of localizations and drifts that sometimes can happen using the laser and vehicle odometry only. After a long testing phase, the system is actually installed in a modern industrial warehouse in Parma in order to finally estimate its robustness and reliability.
Keywords :
mobile robots; object detection; object tracking; optical scanners; path planning; reliability; remotely operated vehicles; robot vision; stereo image processing; 3D monitored area; VisLab solution; advanced LGV navigation; guideline tracking; hanging obstacle detection system; industrial environments; industrial warehouse; laser guided vehicles; laser scanner based system; laser-based systems; long testing phase; reliability; smart vision system; stereo system; unmanned vehicles; vehicle odometry; Cameras; Image edge detection; Lasers; Machine vision; Mathematical model; Navigation; Vehicles;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Intelligent Transportation Systems (ITSC), 2012 15th International IEEE Conference on
Conference_Location :
Anchorage, AK
ISSN :
2153-0009
Print_ISBN :
978-1-4673-3064-0
Electronic_ISBN :
2153-0009
Type :
conf
DOI :
10.1109/ITSC.2012.6338760
Filename :
6338760
Link To Document :
بازگشت