The navigation R package allows to analyze the impact of sensor error modeling on performance of integrated navigation (sensor fusion) based on IMU, GPS (generally speaking, GNSS), and barometer data. The package allows for one of the two major tasks:

• Sensor model evaluation: The user shall provide a reference trajectory, along which the navigation performance is being evaluated using different sensor error models. Perfect sensor data along that reference trajectory are generated, and then corrupted by sensor error coming from either simulation based on the error models provided by user, or directly from user input (option to be added). Integrated navigation is then performed, whit a separately provided error model to be used within the Extended Kalman Filter (EKF). The user can easily introduce GPS outage periods, and there is a growing number of tools to visualize and summarize the results.

• Integrated navigation (sensor fusion) As a natural by-product of the first main application, integrated navigation is also available to users. Providing only the sensor data and the sensor error model to be used within the navigation filter, the user is able to perform integrated navigation using the package and also benefit from a subset of visualization tools.

Caution A flat non-rotating Earth model is assumed throughout the package. We consider this not to be of major impact on sensor model evaluation, as the main contributor there are match/mismatch between the additive sensor errors and the provided error models to the navigation filter. For absolute navigation results though, is long distances and high speeds are involved, such simplifications start to have measurable impact on results. Also, attitude parameterization is done via Euler angles at the moment, bringing their interinsic limitations, such as the singularity at pitch