Nonlinear Techniques for Seamless Low-Cost Portable Navigation in all Environments

This paper targets the development of nonlinear techniques for achieving an accurate, low-cost, always-available navigation solution through integration of Micro-Electro Mechanical Systems (MEMS) sensors with Global Navigation Satellites Systems (GNSS) such as the Global Positioning System (GPS) using nonlinear filtering and nonlinear error-state models. The general objective of this work is the development of nonlinear filtering-based techniques and algorithms to enable navigation of a device that can either move freely without constraints within another moving platform, move with constraints within the moving platform, or is tethered to the moving platform. One of the main applications for this is the navigation of a portable system (for example a cell phone), and the objective for such a system is to work seamlessly in outdoor and indoor environments, in walking or driving modes. The accuracy of this system should not be affected by the change in mode or environment. The used sensors are low-cost MEMS-based inertial sensors (3-axes accelerometers and 3-axes gyroscopes) integrated with GPS during its availability; optionally barometers and 3-axes magnetometers can be used. The approaches that will be used to achieve this are: (1) Utilization of Mixture Particle Filtering (PF) which is a nonlinear filter that can accommodate arbitrary sensor characteristics, motion dynamics and noise distributions; (2) Design of nonlinear error-state system model rather than the approximated linearized error models used in the literature or the nonlinear total-state system model already presented; (3) Extra modules are utilized to detect the current mode of transit (i.e. walking or driving) and to detect the portable device misalignment between the device coordinate frame and the vehicle/person platform coordinate frame. The algorithm was tested and the performance of the proposed navigation system has been verified in real-life scenarios including driving and walking. It showed very competitive performance for such low-cost sensors.