Magnetic sensor operation onboard a UUV: magnetic noise investigation using a total-field gradiometer

To operate a magnetic sensor on board an Unmanned Underwater Vehicle (UUV), it is necessary to provide a means for compensating magnetic noise from a variety of sources. In previous applications involving passively towed platforms, the noise arises from the magnetic fields due to eddy currents and magnetic polarization changes occurring due to rotation of the platform in the earth's magnetic field. It has been demonstrated that this noise can be compensated with a single vector magnetometer that is rigidly attached to the sensor of interest, measuring the rotational changes in the earth's magnetic field. This compensation algorithm has been extended, for passively towed platforms, to provide both motion noise reduction and localization of magnetic dipole targets, in a single process. On a UUV, noise sources are present in addition to those induced by the earth's magnetic field. There are active elements present, such as current loops, motors, actuators, sonars, and electronic devices. To compensate for these additional sources, it is necessary to add additional reference sensors, and to employ more complex compensation models. In a recent MTS/IEEE Oceans paper, a preliminary investigation of such models was made for several prototype UUVs, for both a fluxgate tensor gradiometer and a single-axis total field gradiometer. The total-field gradiometer is the most sensitive of the devices and was operated onboard an active Bluefin BPAUV UUV. The UUV was mounted on a three-axis motion table, and the results obtained were encouraging. In the present paper, we apply a multi-reference sensor magnetic noise compensation algorithm to a single-axis total-field gradiometer operating in the environment of a completely redesigned Bluefin platform, the RELIANT UUV, again mounted on a three-axis motion table. This vehicle has a different configuration from that of the BPAUV, including a very different battery design. We examine in detail the effect of the positioning and number of reference sensors, and show that these issues are critical to performance. We investigate the effect of the positioning of the magnetic sensor relative to other platform systems and find this to be an important issue. Finally we conclude that the performance of the sensor onboard the RELIANT platform using all available reference sensors is roughly the same as that demonstrated onboard the BPAUV.