Robust tracking of a mobile beacon using time differences of arrival with simultaneous calibration of receiver positions

Abstract

Abstract-Localization based on time differences of arrival (TDOA) has turned out to be a promising approach when neither receiver positions nor the positions of signal origins are known a priori. In this paper, we consider calibration-free tracking of a mobile beacon using TDOA, i.e., the positions of the receivers are not given. We propose a probabilistic formulation using a particle filter to simultaneously localize the signal beacon and the receivers. Our method is robust against measurement outliers and incorrect initialization. This is achieved through a probabilistic sensor model for TDOA data which explicitly considers the measurement uncertainty and takes into account disproportional errors caused by measurement outliers. For the reliable initialization of the particle filter, we apply an iterative optimization approach to multiple subsets of TDOA data, where the best solution is implicitly selected by appropriate weighing of the sensor model. We verify the robustness of our approach in extensive experiments in a spacious indoor environment by an ultrasound beacon moving on various trajectories. We demonstrate that our approach ensures a proper initialization of the particle filter and provides accurate position estimates for the signal beacon and the receivers even in case of measurement outliers. Compared to position references of an optical motion capture system we achieve mean position errors below 5 centimeters.

Acoustic Self-Calibrating System for Indoor Smart Phone Tracking

“…The algorithm is described in [32]. Our method is robust against measurement outliers and incorrect initialization.…”

Unsynchronized ultrasound system for TDOA localization

“…The algorithms used are squared or maximum likelihood estimators [17], particle filters [18], [19] or Kalman filters [20], [21]. We consider the inverted scenario where a moving receiver is located.…”

Virtual testbed for indoor localization

“…A big advantage of this approach is that the sensor hardware needed for such an experiment is less expensive and sensors are easily available. Ultrasound based experiments were conducted by Wendeberg et al [9] as well as by Priyantha et al …”