Navigation of an underwater vehicle using AHRS and depth sensor

This paper describes a localization method of an underwater vehicle. The method uses the measurements from Attitude Heading Reference System (AHRS), depth sensor, and Doppler velocity log (DVL). Generally, AHRS has been used for attitude estimation, and the velocity by the DVL is transformed to the velocity in Earth fixed coordinate system using the estimated attitude. Then, the velocity is integrated to calculate the location of an underwater vehicle. This paper proposes a method that uses depth measurement for improvement of the location estimation. The method augments the estimation state to include the depth along with roll, pitch and yaw. EKF approach is used for estimation of attitude and location of the vehicle. Simulation results verify the improvements through fusion of depth measurement. Though depth measurement carries only the information of z coordinate of the robot location, fusing the depth improves localization accuracy in x and y coordinate as well as z coordinate.