Seamless navigation for indoor-outdoor positioning using GNSS-aided UWB/WiFi/IMU system

The need for seamless indoor-outdoor navigation is growing in different application fields, especially for factory scenarios, military, or first response emergency services. Multi-sensor fusion technology has become very prominent for seamless navigation systems owing to its complementary capabilities to Global Navigation Satellite System (GNSS) positioning. Machine learning (ML) and artificial intelligence (AI) solutions have also been widely adopted in literature in the last few years combining them with localization techniques for error mitigation and maximizing overall accuracy and system integrity. The research work of this paper is aimed at the following: firstly, to design an indoor-outdoor (IO) detection strategy that helps to correctly identify the transition between outdoor and indoor with reduced latency; secondly, to construct two separate integration schemes. GNSS integrated with indoor positioning technology (GNSS/UWB/IMU and GNSS/WiFi/IMU) can account for the GNSS signal distortion in indoor and urban environments. Based on this context, a loosely coupled architecture is implemented. The GNSS receiver helps the IMU to update with absolute positions from GNSS in the outdoor scenario. This positioning strategy takes advantage of both GNSS and UWB/IMU/WiFi combinations to realize a seamless indoor-outdoor positioning for personnel and indoor robots moving between the outdoor and indoor environments. The correct detection of the IO transition is essential in seamless IO positioning. This enables making the right decision on the navigation mode. Our implementation of indoor-outdoor (IO) detection strategies makes use of ML to find the IO signal transition pattern. The proposed system would be tested in a scenario over 1.1 km including outdoor, and indoor phases.