Tracking Multiple Moving Objects for Real-Time Robot Navigation

  • Authors:

  • Erwin Prassler;Jens Scholz;Alberto Elfes

  • Affiliations:

  • Research Institute for Applied Knowledge Processing (FAW), P.O. Box 2060, D-89010 Ulm, Germany. prassler@faw.uni-ulm.de;Research Institute for Applied Knowledge Processing (FAW), P.O. Box 2060, D-89010 Ulm, Germany;Automation Institute, Informatics Technology Center (CTI), P.O. Box 6162, 13089-500 Campinas, SP, Brazil. elfes@ia.cti.br

  • Venue:
  • Autonomous Robots
  • Year:
  • 2000

Quantified Score

Hi-index 0.00

Visualization

Abstract

This paper proposes a method for detecting and tracking themotion of a large number of dynamic objects in crowded environments,such as concourses in railway stations or airports, shopping malls,or convention centers. With this motion information, a mobile vehicleis able to navigate autonomously among moving obstacles, operating athigher speeds and using more informed locomotion strategies thatperform better than simple reactive manoeuvering strategies. Unlikemany of the methods for motion detection and tracking discussed inthe literature, our approach is not based on visual imagery but uses2D range data obtained using a laser rangefinder. The directavailability of range information contributes to the real-timeperformance of our approach, which is a primary goal of the project,since the purpose of the vehicle is the transport of humans incrowded areas. Motion detection and tracking of dynamic objects isdone by constructing a sequence of temporal lattice maps. Thesecapture the time-varying nature of the environment, and are denotedas time-stamp maps. A time-stamp map is a projection of rangeinformation obtained over a short interval of time (a scan) onto atwo-dimensional grid, where each cell which coincides with a specificrange value is assigned a time stamp. Based on this representation,we devised two algorithms for motion detection and motiontracking. The approach is very efficient, with a complete cycleinvolving both motion detection and tracking taking 6 ms on a Pentium166 MHz. The system has been demonstrated on an intelligentwheelchair operating in railway stations and convention centersduring rush hour.