Triangulation-based indoor robot localization using extended FIR/Kalman filtering

Abstract

A combined extended finite impulse response (EFIR) and Kalman (EFIR/Kalman) algorithm is proposed for mobile robot localization via triangulation. A distinctive advantage of the EFIR algorithm is that it completely ignores the noise statistics which are typically not well known to the engineer. Instead, it requires an optimal averaging interval of N <inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">opt</inf> points. To run this algorithm, several initial Kalman estimates are used for the roughly set noise covariances. We consider a mobile robot travelling on an indoor floorspace and localized via triangulation with three nodes in a view. We show that the EFIR/Kalman filter is more accurate than the extended Kalman filter under the uncertain noise statistics and initial state.