Novel real-time adaptive control of multiple co-located acoustic sensors for ocean mapping and other hydrographical applications

In this thesis a novel approach to the management of multiple payload sonar systems is described. The objective of this study is the design, development, and experimental verification of an automated system that facilitates the close-proximity integration and concurrent operation of multiple similar high-frequency acoustic sensors. The increasing emphasis in the collection and fusing of multiple datasets of a region of the seafloor is highlighted, providing evidence that the provision, and combined analysis, of complementary datasets, affords a more accurate representation of the seafloor, the removal of possible dataset ambiguities and improved data analysis and interpretation. A literature review of current state-of-the-art in payload sensor management and UUV payload integration identifies a number of performance and design shortfalls. By current techniques, UUV-based payload sensor control management is still very much operator-dependent, employing fixed or pre-deployment programmed routines. Recent advances in computational technology and real-time programming techniques afford the ability to process bathymetric data in-situ. The novel approach of this thesis interrogates real-time bathymetric data to predict the transmission-reception timing of payload sensor acoustic pulses, thus permitting the ability to synchronise the trigger of the instruments such that neighbouring return signals of other sonar are not saturated. The new approach to the automated control of payload sonar is experimentally tested and results show it to offer enhanced sensor deployment and increased survey efficiency through the control of sounding density and provides for platform velocity optimisation, throughout an entire survey. Most importantly, it is shown that the automated system permits the deployment of multiple high-frequency acoustic sensors concurrently operating in close-proximity to the seafloor, without the necessity of separating the operational frequencies compromising the resolution of the acquired datasets.