A Preliminary Study on Attitude Measurement Systems Based on Low Cost Sensors

The increasingly use of Autonomous Underwater Vehicles (AUVs) in several context led to a rapid development and enhancement of their technologies, allowing the automatization of many tasks. One of the most challenging tasks of AUVs still remains their robust positioning and navigation, since classical global positioning techniques are generally not available for their operations. Inertial Navigation System (INS) methods provide the vehicle current position and orientation integrating data acquired by the internal accelerometer and gyroscope. This system has the advantage of not needing to either send or receive signals from other systems; however, among the errors the sensors are mainly affected by, the most critical one is related to their drift, which makes the position error growing over time. The attenuation of the effect of these problematics is generally achieved combining different positioning methods, as for example acoustic- or geophysical-based ones. An accurate estimation of the device orientation is anyway necessary to get satisfying results in terms of position and autonomous navigation. In this paper, a preliminary study on the use of smartphone low-cost sensors to perform attitude estimation is presented. With the final aim of developing a cheaper and more accessible underwater positioning system, a first analysis is conducted to verify the accuracy of the attitude angles obtained by the integration of smartphone data acquired in different operative settings. Different filtering methods will be employed.