Radar Localization with multiple Unmanned Aerial Vehicles using Support Vector Regression

Author

Sundaram, Bharat ; Palaniswami, Marimuthu ; Reddy, Sanjay ; Sinickas, Michael

Author_Institution

Dept. of Electr. & Electron. Eng., Melbourne Univ., Vic.

fYear

2005

Firstpage

232

Lastpage

237

Abstract

This paper presents a first attempt to solve the geolocation problem using support vector regression (SVR). This paper proposes a method to pinpoint the location of stationary, hostile radar using the time difference of arrival (TDoA) of the same characteristic pulse emitted by the radar at 3 different unmanned aerial vehicles (UAVs) flying in a fixed triangular formation. The performance of the proposed SVR method is compared with a variation of the Taylor series method (TSM) used for solving the same problem and currently deployed by the DSTO, Australia on the Aerosonde Mark III UAVs. The robustness to error of the SVR method is explored and compared with the TSM. Extended applications of the SVR approach to more general localization scenarios in wireless sensor networks are proposed for further work

Keywords

aerospace robotics; aircraft; mobile robots; radar tracking; remotely operated vehicles; series (mathematics); support vector machines; wireless sensor networks; Taylor series method; geolocation problem; multiple unmanned aerial vehicles; radar localization; support vector regression; time difference of arrival; wireless sensor networks; Australia; Monitoring; Radar tracking; Robustness; Signal processing; Target tracking; Taylor series; Time difference of arrival; Unmanned aerial vehicles; Wireless sensor networks;

fLanguage

English

Publisher

ieee

Conference_Titel

Intelligent Sensing and Information Processing, 2005. ICISIP 2005. Third International Conference on

Conference_Location

Bangalore

Print_ISBN

0-7803-9588-3

Type

conf

DOI

10.1109/ICISIP.2005.1619441

Filename

1619441