SpringerOpen Newsletter

Receive periodic news and updates relating to SpringerOpen.

This article is part of the series Cooperative Localization in Wireless Ad Hoc and Sensor Networks.

Open Access Research Article

Tracking Objects with Networked Scattered Directional Sensors

Kurt Plarre1 and P R Kumar2*

Author Affiliations

1 Department of Mechanical Engineering and Center for Control, Dynamical Systems and Computation, University of California, Santa Barbara, CA 93106, USA

2 Department of Electrical and Computer Engineering and Coordinated Science Laboratory, University of Illinois at Urbana-Champaign, 1308 W. Main St., Urbana, IL 61801, USA

For all author emails, please log on.

EURASIP Journal on Advances in Signal Processing 2008, 2008:360912  doi:10.1155/2008/360912


The electronic version of this article is the complete one and can be found online at: http://asp.eurasipjournals.com/content/2008/1/360912


Received:19 April 2007
Accepted:4 August 2007
Published:9 August 2007

© 2008 The Author(s).

This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Abstract

We study the problem of object tracking using highly directional sensors—sensors whose field of vision is a line or a line segment. A network of such sensors monitors a certain region of the plane. Sporadically, objects moving in straight lines and at a constant speed cross the region. A sensor detects an object when it crosses its line of sight, and records the time of the detection. No distance or angle measurements are available. The task of the sensors is to estimate the directions and speeds of the objects, and the sensor lines, which are unknown a priori. This estimation problem involves the minimization of a highly nonconvex cost function. To overcome this difficulty, we introduce an algorithm, which we call "adaptive basis algorithm." This algorithm is divided into three phases: in the first phase, the algorithm is initialized using data from six sensors and four objects; in the second phase, the estimates are updated as data from more sensors and objects are incorporated. The third phase is an optional coordinated transformation. The estimation is done in an "ad-hoc" coordinate system, which we call "adaptive coordinate system." When more information is available, for example, the location of six sensors, the estimates can be transformed to the "real-world" coordinate system. This constitutes the third phase.

Publisher note

To access the full article, please see PDF.