Research On Underwater Object Detection Using A Femtosecond Laser Imaging LiDAR

With the rapid development of science and technology,people have become increasingly aware of the importance of the ocean for the national economy and homeland security.Among the many areas of technology,as a high accuracy and efficiency tool of underwater perception,laser imaging detection has an important role in supporting the national development strategy of maritime power.The thesis encompasses the demand for underwater precision measurements,uses the light source characteristics as the breakthrough point,and proposes an underwater LiDAR imaging mode that combines femtosecond laser interferometry and galvanometer scanning.The system utilizes the superiority of femtosecond optical frequency comb for metrology and detection,innovates the sensing mechanism of underwater pulse laser,and achieves integrated optimization of point cloud imaging mode via the system-on-a-chip.The main research contents are as follows:(1)Analyze the time-frequency domain metrology characteristics of the femtosecond optical frequency comb and the underwater law of transmission and detection.The thesis expounds on the preparation mechanism of fiber mode-locked laser,which locks the repetition frequency using the phase-locking loop.Based on the interference detection mode of femtosecond laser in the time-frequency domain,the extraction of distance cues in echo signal is analyzed and derived.The underwater attenuation model of the femtosecond laser is analyzed,and the LiDAR equation is established following the underwater laser detection law.Besides,the benefits of the light source in environment adaptation and detection range are discussed.(2)Study the underwater absolute distance measurement of the optical frequency comb.The thesis proposes an underwater ranging technology based on the chirp accumulation from the water,and the accuracy is optimized with the phase of synthetic wavelength,and measuring uncertainty can be well within 4 μm.The thesis proposes an underwater absolute distance measurement method,which utilizes a dynamic frequency comb and chirped pulse interferometry,the detection performance is enhanced by long fiber delay lines and the charge-coupled device,and the ranging accuracy is below 21 μm.The thesis proposes a cross-sampling dual-comb ranging method,which combines the correction module of the water group refractive index.It improves the data update rate by the water dispersion while obtaining the ranging measurement uncertainty of 25.5 μm.(3)Propose the scanning imaging model of the point cloud for femtosecond LiDAR.The thesis introduces the galvanometer driving model and the scanning drive modes.The accurate matching of feedback angle and distance is completed under the timing sequence.Based on the imaging coordinate system,the spatial reduction model of underwater imaging LiDAR is established,and the denoising optimization of detection data is realized using the point cloud filtering algorithms.Finally,the pixel display of point cloud imaging results is performed by gridding the data.(4)Perform the system development of underwater femtosecond laser imaging LiDAR.The thesis constructs the underwater point cloud imaging system by dualcomb interferometry combined with galvanometer scanning,and the optical receiving module is designed according to the LiDAR equation and the coherence condition.Peripheral hardware is designed considering functional requirements,and integrated implementation for LiDAR is realized based on a system-on-a-chip development board.The system determines the structure size and device parameters,and assembly and joint debugging of the prototype are performed.The imaging experiments of the laboratory and the testing pool in the National Ocean Technology Center are completed,and the results determine the performance parameters of the prototype system.Finally,the accuracy of imaging detection is better than 180 μm.