Indoor Floor Detection and Localization Based on Deep Learning and Particle Filter

In this chapter, we present an infrastructure-independent multi-floor indoor localization scheme. The proposed scheme has several notable features. First, we utilize the strong feature extraction capability of the sequence-to-sequence (Seq2Seq) deep learning model for sequential data to implement real-time step action prediction. We also develop a floor decision algorithm to extract vertical movement information from the step action sequence under a variety of user activities. Second, we configure calibration nodes on the map based on prior knowledge from the environmental information to extend the localization to three-dimensional applications and achieve calibration of the estimation. Third, we introduce a clustering method to improve localization performance in uncertain measurements. The experimental results show that the Seq2Seq model has good robustness to noisy data. Under the long path multi-floor scenario, our scheme achieved a localization accuracy of over 96% within a 2 m error boundary.