Hardware Design Of Forward-looking Sonar Data Processing System Based On Embedded GPU

With the continuous development of exploration of marine resources,deep-sea resource exploration has become a hot research topic at home and abroad.Forward-looking sonar system,as the key equipment for obstacle avoidance of deep-sea submarines,has received extensive attention.Forward-looking sonar actively emits the acoustic beam and calculates the echo signal to obtain the underwater acoustic image,which has a very broad application prospects in underwater vehicle obstacle avoidance,deep-sea exploration and salvage,seabed topographic mapping and other fields.At present,the research of forward-looking sonar system for deep-sea submarine obstacle avoidance is still immature at home.Most of them rely on imports,which is expensive and difficult to customize the design according to the actual applicationFor this reason,this thesis studies and develops the data processing hardware system of forward-looking sonar with independent intellectual property rights.The system takes embedded GPU Tegra K1 as the core,realizes data acquisition,data processing,data caching and transmission functions.The system filters and amplifies the signals collected by the sensor array,uses FPGA to sample,cache and DFT the acoustic data,then uploads the processing results and sampling data to embedded GPU through PCIe bus.The embedded GPU calculates the received data by beamforming,and uploads the calculation results to the host computer through Gigabit Ethernet for sorting and displaying.In this thesis,the signal integrity of the hardware system is simulated and verified,and the power supply and key signal quality of the system are tested.The analog signal indicators and pulse emission are tested.The interface function verification and oil pressure experiment are completed,and the overall function of the prototype is tested to verify the validity and reliability of the system.