IMU alignment for smartphone-based automotive navigation

Recent years have seen an increasing interest in making use of the smartphone as a cheap and viable navigation device for land-vehicles. However, smartphone-based automotive navigation suffers from the fact that the orientation of the smartphone's inertial measurement unit, with respect to the vehicle, in general is unknown. In this study, we present a method for simultaneous vehicle navigation and smartphone-to-vehicle alignment. In addition to the state estimates obtained from applying a standard global navigation satellite system-aided inertial navigation system to estimate the smartphone dynamics, this will also provide us with estimates of the vehicle's attitude. These estimates are used to improve the navigation solution, and also enables the estimation of additional vehicle, road, and driver characteristics, which requires knowledge of the vehicle's attitude. The performance of the proposed method is evaluated with both simulations and experimental data.