Indoor PDR performance enhancement using minimal map information and particle filters

摘要: For professional users such as firefighters and other first responders, GNSS positioning technology (GPS, assisted GPS) can satisfy outdoor requirements in many instances. However, there is still a need for high-performance deep indoor use by these same users. This has already been clearly expressed various communities of end the context WearIT@Work, an R&D project funded European Community's Sixth Framework Program. It known that map matching help pedestrian navigation. In most previous research, it was assumed detailed building plans are available. emergency / rescue scenarios, only very limited plan information may be at hand. example outline might obtained from aerial photographs or cataster databases. Alternatively, escape posted entrances to would yield approximate exit door stairwell locations well hallway room orientation. What not how much really required USAR mission each level detail improve accuracy. Obviously, geometry course through will factors consider. The purpose this paper show previously published Backtracking Particle Filter (BPF) combined with different levels PDR performance. A new in/out scenario typical reconnaissance during fire two-story office evaluated. Using external wall information, yields performance (2.56 m mean 2D error) greatly superior PDR-only, no base case (7.74 error). result substantial practical significance since could quickly easily generated technique used mitigate heading errors exposing IMU extreme operating conditions. hoped mitigating effect also occur more irregular paths larger traversed spaces parking garages warehouses.