A Sensor System for the Navigation of an Underwater Vehicle

Abstract

A sensor system for an underwater vehicle is described. The vehicle is equipped with inclinometers, gyroscopes, a magnetometer, a pressure gauge, and a sonar system. The sensor models used for the inclinometers and gyroscopes are straightforward; however, the magnetometer can be corrupted by variations in the earth’s field caused by: external objects and internal magnetic fields. We show how to use inclinometer data to adjust for a limited set of external field variation. We also show how to calibrate the magnetometer to compensate for static and thruster-dependent internal fields. The sonar unit uses range differentials between cheap time-of-flight sonar to follow a target. This reduces signal processing since data association is only required on target acquisition, and removes the need to scan an entire landscape, which is usually slow. The gyroscopes are fused via a second indirect filter system. The vehicle attitude is represented as a quaternion; these have a low computational burden, and lack discontinuities and singularities. The simplicity of the indirect filter permits very fast update rates, so that the system may follow rapid vehicle rotations.