A Reference Measurement Data Set for Multisensor Pedestrian Navigation with Accurate Ground Truth

This paper presents a measurement dataset for testing and evaluating multi-sensor approaches in pedestrian navigation. The measurements include both transitions from outdoor to indoor and vice versa. Furthermore, segments with explicit threedimensional character, such as ramps, stairs and elevators are included. The measurements have been carried out in and around a lab and office building. Ground truth reference points are provided with sub-centimeter accuracy. The results of two implementations of Bayesian filter algorithms (EKF and Particle filter) are provided for reference purposes. All measurements and background data are publicly available via an accompanying website.