Tracking of Moving Objects from a Moving Vehicle Using a Scanning Laser Rangefinder

Author

MacLachlan, Robert A. ; Mertz, Christoph

Author_Institution

Robotics Inst., Carnegie Mellon Univ., Pittsburgh, PA

fYear

2006

fDate

17-20 Sept. 2006

Firstpage

301

Lastpage

306

Abstract

The capability to use a moving sensor to detect moving objects and predict their future path enables both collision warning systems and autonomous navigation. This paper describes a system that combines linear feature extraction, tracking and a motion evaluator to accurately estimate motion of vehicles and pedestrians with a low rate of false motion reports. The tracker was used in a prototype collision warning system that was tested on two transit buses during 7000 km of regular passenger service

Keywords

collision avoidance; feature extraction; laser ranging; motion estimation; navigation; object detection; optical tracking; vehicle dynamics; 7000 km; autonomous navigation; collision warning system; future path prediction; linear feature extraction; motion evaluator; moving object detection; moving object tracking; moving vehicle sensor; pedestrian motion estimation; scanning laser rangefinder; vehicle motion estimation; Alarm systems; Feature extraction; Motion estimation; Navigation; Object detection; Prototypes; Remotely operated vehicles; Sensor systems; System testing; Tracking;

fLanguage

English

Publisher

ieee

Conference_Titel

Intelligent Transportation Systems Conference, 2006. ITSC '06. IEEE

Conference_Location

Toronto, Ont.

Print_ISBN

1-4244-0093-7

Electronic_ISBN

1-4244-0094-5

Type

conf

DOI

10.1109/ITSC.2006.1706758

Filename

1706758