Interactive simulation of IMU, Doppler surrogate, and acoustic beacon fusion with Extended Kalman Filter state estimation and real-time drift visualization.
The navigation system employs an Extended Kalman Filter (EKF) with a 15-state vector including position, velocity, attitude, IMU biases, and DVL scale factor.
IMU drift follows a random walk model where position error grows quadratically with time in the absence of aiding sensors:
Where σa is accelerometer noise density and σg is gyroscope noise density. DVL aiding bounds velocity errors, while acoustic fixes provide absolute position corrections.