| MEMS micromirror pulsed Li DAR has become one of the mainstream choice of ve-hicle environment sensing Li DAR at this stage due to its strong directivity,narrow beam,simple system architecture and easy miniaturization.At present,MEMS micromirror vehicle borne pulsed Li DAR has some problems,such as low human eye safety level,limited beam scanning range,low point cloud imaging data rate and poor ranging perfor-mance,which need to be improved urgently.This paper focuses on the key science and technology of signal processing for multi-beam MEMS micromirror vehicle borne pulsed Li DAR,and analyzes the problems of 1550 nm laser pulse transmission and reception with high data rate,walking error suppression,high-precision time interval measurement method,and target detection under low signal-to-noise ratio.The main research contents of this paper are as follows:（1）The method and strategy of multi-beam high repetition frequency Li DAR trans-mitting and receiving.The array grating scanning pattern is designed by combining multi-beam laser sources and MEMS micromirrors to achieve a scanning frame rate of11.76 Hz and a large scanning field of view of about 48°×26°.A photodetector module design based on the four quadrant In Ga As avalanche photodiode is proposed,which meets the actual requirements of large receiving aperture and high bandwidth of the pulsed Li-DAR.Aiming at the problem that multi-beam pulse laser echo signals are easy to mix with each other,the multi signals conditioning scheme based on time division multiplexing is proposed.By sequentially triggering laser source multiplexing coding and multiplexer se-lection switch demultiplexing,the orderly single channel transmission of 416.667k groups of laser pulse signals per second is realized,which makes it possible for a set of laser pulse timing and measurement system to measure multi-beam laser ranging signals and simplifies the overall circuit.（2）High precision laser ranging signal processing and realization of pulse width compensation.Through analyzing the waveform characteristics,the idea of optimizing pulse timing by using pulse width compensation is introduced,and simulation experiments are carried out.The corresponding relationship between echo pulse width data and echo pulse intensity is established theoretically.Based on the single threshold comparator and TDC-GPX2 chip,the measurement circuits of pulse front time interval and echo pulse width are designed.By using specific compensation method and Kalman filter,the high precision timing and time measurement of Li DAR pulse signal are completed.Through static single point ranging experiment,it is proved that the walking error time value of laser pulse time of flight measurement with pulse width compensation method is less than100 ps,that is,the walking error is less than 15 mm.（3）According to the theoretical analysis and system design,the principle prototype is built,and the experiment of high data rate MEMS Li DAR time-division multiplexing pulse width compensation imaging system is completed.A frame synchronization mechanism by sending data are designed.The space Cartesian rectangular coordinate transformation matrix of prototype data is established,and the comparator threshold is set according to the detection probability curve.Through the indoor and outdoor experiments of the Li DAR prototype,it is proved that the prototype achieves the data rate index of 308.7 s-1deg-2,the three-dimensional target reconstruction error of less than 5%and the detection range of about 138 m.（4）High precision signal processing of full waveform pulse centroid timing algo-rithm.An improved heavy-tailed function model is proposed to describe the time domain distribution of Li DAR pulse waveforms.In which,a full waveform centroid algorithm based on pulse machine learning is proposed.In this algorithm,the high amplitude pulse waveform is obtained by slope screening,the quadratic curve is fitted,and the centroid of the waveform is determined by machine learning.Simulation and experimental results show that the algorithm has good ranging performance under normal signal-to-noise ratio.A centroid algorithm of Li DAR pulse based on improved Empirical Mode Decomposi-tion and intensity weighting is proposed.The improved Empirical Mode Decomposition adaptively filters the high-frequency noise of the pulse signal,and the centroid of the pulse waveform is obtained by using the intensity weighted calculation.The experimental results show that the algorithm still has good ranging performance under a extremely low signal-to-noise ratio. |
PDF Full Text Request