An audio-based strategy for the geo-referencing of visual data to support diver's monitoring missions

The collaboration between humans and robots represents a major topic in robotics. In the underwater domain, the employment of unmanned robots could help divers perform number of tasks such as monitoring and inspection activities. However, the underwater environment poses severe limitations that still need to be overcome to achieve autonomous operations. This research work represents an intermediate step in the human-robot switchover in underwater domain for a particular application and involves the use of ICT tools to facilitate human divers in the monitoring of Posidonia Oceanica (Po) meadows. The paper reports a strategy to geo-reference underwater visual data exploiting the synchronization between the audio track recorded by a camera and the transponder pings adopted for the acoustic positioning system. The paper also reports the exploitation of the geo-referenced data for the identification of Po meadow over the mission area using a Machine Learning algorithm. The results are very promising and can lead to an accurate geo-referenced identification of Po and the reconstruction of the surveyed area.