Smartphone-based GNSS/PDR integration navigation enhanced by measurements resilient adjustment under challenging scenarios

The integration of Global Satellite Navigation System (GNSS) and Pedestrian Dead Reckoning (PDR) is one of the widely adopted solutions for smartphone-based pedestrian navigation. Due to the effects of multipath and Non-Line of Sight (NLOS) signals, low-cost GNSS chips embedded in smartphones face significant challenges in complex environments such as urban canyons, thereby threatening the availability of positioning information in integrated navigation systems. To address this problem, a GNSS/PDR integration navigation method enhanced by measurements resilient adjustment is proposed. Initially, an augmented GNSS/PDR model is established that leverages GNSS velocity to estimate step length, which serves as an additional GNSS measurement. Due to the poor measurement redundancy, GNSS measurements are then categorized and processed accordingly in the loosely-coupled model. For attaining position solutions, a resilient adjustment algorithm assisted by fault repair is proposed to correct GNSS measurements. For step length measurements, a step length test statistic is constructed, and anomalous step lengths estimated with GNSS are appropriately excluded. Experimental results demonstrate that the proposed method can effectively utilize GNSS observational data and perform targeted processing, thereby significantly enhancing the availability and continuity of positioning information in pedestrian navigation.