Sigma-point Kalman smoothing for indoor tracking and auto-calibration using time-of-flight ranging

Reliably estimating the location of people and tracking them in an indoor environment poses a fundamental challenge for existing commercial tracking systems. We are currently developing a real time indoor location tracking system specifically for long-term monitoring of patients in a health care setting. The development is in collaboration between EmbedRF LLC and Oregon Health & Science University (OHSU). This paper focuses on the algorithmic methods being developed that are also applicable to a broad range of pedestrian monitoring and ubiquitous computing applications. At the core of our system is a sigma-point Kalman smoother (SPKS) based Bayesian inference approach. Time-of-flight (TOF) range measurements from multiple access points are fused with a model of human walking to determine a person's 2D position and velocity. The SPKS also performs "auto-calibration" or simultaneous localization and mapping (SLAM) to determine scaling, offset, and the 2D location of the wall-mounted access points. The indoor tracking accuracy of the proposed system is better than 1 meter (m).