Coaxial scanning type ultra-long distance detection laser radar system

Abstract

The invention relates to the technical field of target distance detection and 3D scanning reconstruction, and discloses a coaxial scanning type ultra-long distance detection laser radar system, which comprises a laser light source module, which adopts a 1550nm optical fiber laser to output high-power laser, and ensures sufficient laser energy and human eye safety; according to the invention, by adopting the high-power 1550nm fiber laser, the ultra-long distance detection is realized on the premise of ensuring the safety of human eyes, and the detection range of more than 10km can be reached, so that the limitation that the detection distance of the traditional laser radar is only 500m is broken; according to the invention, by selecting the 1550nm laser device which is safer to human eyes and greatly improving the laser power to kilowatt and even myriawatt levels, stronger echo signals are realized, so that the detection signal-to-noise ratio is remarkably improved; the micro-lens array is adopted to replace a traditional DOE diffractive optical element, efficient beam shaping of a one-dimensional or two-dimensional dot matrix is achieved through flexible regulation and control of the curvature radius and the distance of the micro-lenses, meanwhile, the manufacturing cost is reduced, and the design freedom degree is expanded.

Description

Technical Field

[0001] The present invention relates to the technical fields of target distance detection and 3D scanning reconstruction, and specifically provides a coaxial scanning ultra-long distance detection lidar system. Background Art

[0002] The coaxial scanning ultra-long distance detection lidar system is a lidar system that adopts a coaxial optical path design. By emitting laser pulses and receiving the signals reflected by the target, it realizes ultra-long distance target detection and three-dimensional shape construction. It has the advantages of no blind area, good angle consistency, and good linearity of energy with distance, and has broad application prospects in the fields of autonomous driving, topographic mapping, environmental monitoring, etc.

[0003] However, in the prior art, there are the following defects and deficiencies:

[0004] 1. Existing lidars can currently only detect distances within 500m. For detections beyond 500m, the reflected light signals received by the receiver will attenuate sharply.

[0005] 2. Currently, the mainstream uses 905nm near-infrared laser light sources, which are prone to cause potential harm to the human eye under high-power output, and the signals received during ultra-long distance detection are extremely weak and cannot meet the requirements.

[0006] 3. Traditional lidars usually adopt a relatively large field of view (FOV), which is suitable for short-distance large-range detection. However, in ultra-high-speed ultra-long distance detection, a low FOV is more applicable.

[0007] 4. The extremely low FOV angle in ultra-high-speed ultra-long distance detection places high demands on the angular resolution. The angular resolution determines the minimum angular resolution of the selected galvanometer, and at the same time, a certain laser beam expansion ratio still needs to be compatible to avoid the problem of beam divergence in ultra-long distance detection.

[0008] 5. Short-distance detection usually requires a projection lens to project the light emitted by MEMS or array laser diodes to a long distance, while ultra-long distance detection requires a completely different lens design.

[0009] 6. The power of existing laser light sources is low and it is difficult to meet the high-power requirements in low-FOV ultra-long distance detection.

[0010] 7. Traditional lidars usually adopt DOE diffractive optical elements for beam shaping, which have high manufacturing costs and processing difficulties and are not suitable for large-scale applications.

[0011] Therefore, those skilled in the art provide a coaxial scanning ultra-long distance detection lidar system to solve the above-mentioned problems. Summary of the Invention

[0012] In view of the shortcomings of the existing technology, the present invention provides a coaxial scanning ultra-long-range detection laser radar system to solve the problems raised in the above background technology.

[0013] To achieve the above objectives, the present invention is implemented through the following technical solutions: a coaxial scanning ultra-long-range detection laser radar system, comprising:

[0014] The laser light source module uses a 1550nm fiber laser to output high-power laser, ensuring sufficient laser energy and eye safety;

[0015] The beam shaping module converts the output laser into a uniformly directional beam using a microlens array, controlling the divergence angle and maintaining high energy density;

[0016] The scanning beam expansion lens module dynamically adjusts the laser emission direction according to the directional light beam to achieve long-distance coverage in a low field of view;

[0017] The transmitting and receiving optical path modules adopt a coaxial design based on long-distance coverage, and separate the laser emission and echo signals through a beam splitter prism;

[0018] The signal acquisition and processing module, based on the separation of laser emission and echo signals, uses a highly sensitive detector and a high-speed acquisition card to amplify, acquire, and digitize the echo signals, and uses an algorithm to achieve high-precision reconstruction of distance and three-dimensional information;

[0019] Preferably, the laser light source module includes:

[0020] The 1550nm fiber laser unit is responsible for generating high-power laser output to ensure sufficient laser energy, which is suitable for long-distance detection. In the fiber laser, the calculation formula for the laser amplification process gain is:

[0021] G=exp[σ em N2L-σ abs N1L],

[0022] Among them, σ em represents the emission cross section, σ abs represents the absorption cross section, N2 and N1 represent the excited state and ground state atomic number densities respectively, and L represents the effective length of the optical fiber;

[0023] At the same time, in order to accurately simulate the energy level dynamics in the laser, the rate equation is used:

[0024]

[0025] Where τ represents the laser lifetime, A eff represents the effective area, hν represents the photon energy, P pumpRepresents the pump light power, N total Represents the total number density of doped ions in the laser medium, P signal Represents the signal light power Represents the number of photons contained in the pump light per unit time Represents the rate at which ions in the excited state return to the ground state through spontaneous emission or non-radiative transitions Represents the number of photons in the signal light per unit time

[0026] Laser power adjustment unit: Detects the laser output power through a high-speed photodetector. According to the error between the monitored value and the preset target, an adaptive fuzzy PID control algorithm is adopted to achieve the hybrid of traditional PID and fuzzy control. The formula form is:

[0027]

[0028] Among them, u(t) represents the output control quantity, e(t) represents the power error, K p 、K i 、K d Represent the proportional, integral, and differential gains respectively, α represents the weight coefficient, u fuzzy (t) is the control quantity output based on fuzzy rules Represents the time integral of the error Represents the rate of change of the error, u fuzzy (t) represents the output of the fuzzy controller, (1 - α) represents the weighted coefficient complementary to α

[0029] Preferably, the laser light source module further includes:

[0030] Stability control unit, which collects laser output fluctuation data in real time, performs statistical and spectral analysis on the data to determine the fluctuation components, and realizes output stability by adjusting parameters such as pump current, temperature control, and fiber bending compensation according to the fluctuation characteristics. Therefore, a stability control strategy based on model predictive control is adopted, and its optimization problem expression is:

[0031]

[0032] Among them, A represents the prediction time domain length, y(k) represents the predicted output, y ref (k) represents the reference output value, Q and R are the weight matrices of the state error and the control input respectively, u(k) represents the control variable And Represents the quadratic norm with a weighted matrix, Q represents the output error weighted matrix, R represents the control input weighted matrix Represents the measurement of the energy consumption of the control input at time k It represents the measure of the deviation between the system output and the reference value at time k, and is weighted by the matrix Q. Indicates the accumulation of all errors and control costs from the current moment to the next A-1 steps;

[0033] The eye safety monitoring unit uses a high-speed photodetector to monitor the laser output power and compares the real-time detection value with the preset eye safety threshold. When the output power exceeds the safety threshold, the power is automatically shut down. The algorithm formula is:

[0034]

[0035] Among them, S(P) represents the output of the logic function, P represents the current real-time detected laser output power, and P safe represents the preset human eye safety threshold, β represents the steepness adjustment factor; when P is lower than P safe When S(P) is close to 0, it indicates low risk; when P exceeds P safe When , S(P) quickly approaches 1, indicating that the risk increases sharply and triggers a safety protection response;

[0036] To prevent the system from repeatedly triggering or canceling protection due to small fluctuations within the critical range, a hysteresis comparator can be introduced. Its logic formula is:

[0037]

[0038] Among them, S t Indicates the final safety trigger state signal, θ upper and θ lower Represent the upper and lower thresholds of safety trigger and release trigger respectively, S prev Indicates the security status at the previous moment;

[0039] When the calculated risk coefficient S(P) exceeds the upper threshold θ upper When S(P) drops to the lower threshold θ lower When the value is below the threshold, the system will be released from the safe state. When the value is between the two thresholds, the previous state will be maintained to avoid frequent switching.

[0040] Preferably, the beam shaping module includes:

[0041] The microlens array unit converts the original laser into a uniformly directional beam by precisely designing the arrangement and shape of the microlenses. It uses an algorithm based on Zernike polynomial expansion and minimum mean square error optimization to optimize the beam phase and maximize the uniformity of the beam, thereby providing good beam quality for long-distance transmission and high-precision detection. The algorithm formula is:

[0042]

[0043] Among them, I target (x, c) represents the intensity distribution of the target uniform light spot, B(x, c) represents the amplitude distribution of the original laser beam, Z i (x, c) represents the i-th Zernike polynomial, which is used to represent the phase distortion of the beam, a i represents the Zernike coefficient to be optimized, representing the phase compensation parameter of each microlens, F represents the Fourier transform operator, F{·} represents a certain optical transmission operator, |·| represents taking the amplitude of the complex field, [·] 2 represents the quadratic loss, and D represents the number of basis functions used to expand the phase;

[0044] The divergence angle control unit, from beam parameter acquisition, error calculation, optical element adjustment to closed-loop feedback correction, adopts a hybrid optimization strategy of adaptive fuzzy control and genetic algorithm to achieve more precise divergence angle regulation in a complex environment. Its algorithm formula is:

[0045] Δθ = b(θ target - θ measured ) + (1 - b)·GA(p),

[0046] Among them, Δθ is the required divergence angle adjustment amount, θ target and θ measured respectively represent the target divergence angle and the actually measured divergence angle, b represents the weight between linear compensation and non-linear genetic algorithm optimization, and GA(p) represents the non-linear compensation term obtained by using the genetic algorithm (GA).

[0047] Preferably, the beam shaping module further includes:

[0048] The beam uniformity adjustment unit optimizes the beam energy distribution and improves the subsequent detection accuracy. The specific steps include:

[0049] Step 1.1 Use a high-resolution CCD to collect the intensity distribution of the beam cross-section;

[0050] Step 1.2 Divide the light spot area into several sub-regions, calculate the energy proportion and overall mean in each region, and evaluate the uniformity of the beam energy distribution through Shannon entropy;

[0051] Step 1.3 Use a spatial light modulator to regulate the local phase and amplitude of the beam, and continuously update the control parameters according to the real-time feedback;

[0052] Step 1.4 Continuously compare the actually output beam with the preset target distribution through a closed-loop feedback mechanism, and iteratively optimize the adjustment parameters;

[0053] Step 1.5 Construct the combined objective function and iteratively optimize the control parameter vector u of the SLM:

[0054]

[0055] where p i (u) represents the energy proportion in each region, the energy proportion in the i-th sub-region, M represents the total number of sub-regions, μ(I(u)) represents the mean value of the beam energy under the parameter u, and μ i represents the target mean value, λ represents the mean deviation compensation factor, and ln(p target (u)) represents the natural logarithm of p i (u). i (u).

[0056] Preferably, the scanning beam expander lens module includes:

[0057] A afocal beam expander lens unit that expands the laser beam afocally to cover a wider target area while ensuring that the output spot has a uniform energy distribution and low diffraction loss. The specific steps are as follows:

[0058] Step 2.1 Use a high-precision CCD to collect the original laser light field data, including amplitude and phase information;

[0059] Step 2.11 Determine the beam expansion ratio M according to the actual application requirements, and design the target output light field I target (x, y), requiring that the expanded spot has a uniform energy distribution and low diffraction loss while maintaining the overall energy stability;

[0060] Step 2.12 Use a spatial light modulator phase control element to apply a preset phase modulation to the incident light field to achieve afocal beam expansion;

[0061] Step 2.13 To simultaneously optimize the beam expansion matching effect and reduce the influence of diffraction and aberration, introduce a combined optimization objective function and adjust the control parameter vector u to make the system reach an overall optimal state:

[0062]

[0063] where E in (x, y) represents the complex amplitude distribution of the original laser light field, e jφ(x,y,u) represents the phase control function controlled by the parameter vector u, represents the operator describing the propagation of the light field in the afocal beam expansion system, represents the ideal output intensity distribution obtained by scaling the coordinates of the target light field by the beam expansion ratio M, represents the regularization term, and γ represents the regularization weight factor;

[0064] The dynamic scanning control unit adjusts the laser emission direction in real time and supports long-distance detection in a low field of view. The specific steps are as follows:

[0065] Step 2.2 Use a sensor to collect the spatial distribution and propagation path data of the laser beam in real time, and update the beam state x using Bayesian inference. The formula is: k as follows:

[0066] where z k represents the real-time measurement data, P(z k ) represents the prior probability of the observation z k , P(x k |x k-1 ) represents the state transition probability from the previous state x k-1 to the current state x k , P(z k |x k ) represents the likelihood probability of the observation z k occurring given the state x k , and P(x k |z k ) represents the posterior probability distribution of the system state x k after the given observation z k ;

[0067] Step 2.21 Calculate the relative position between the laser beam and the target area based on the prediction data, and quickly achieve preliminary direction correction through pre-adjustment. The algorithm formula is:

[0068] where represents the average value of the entire data, x i represents the estimated position of the i-th sampling, and E represents the number of samplings;

[0069] Step 2.22 Construct a closed-loop feedback system to monitor the deviation between the output beam and the target in real time, and dynamically adjust the scanning control parameters according to the error feedback. The algorithm formula is:

[0070] where e represents the direction error, F represents a positive definite matrix, and the system is stably regulated by minimizing V(e), represents the coefficient.

[0071] Preferably, the scanning beam expander lens module further includes:

[0072] An angle control unit that precisely controls the scanning angle to ensure that the detection range and resolution meet the requirements. The specific steps include:

[0073] Step 2.3 Collect the current laser emission angle data through a high-resolution angle sensor, and perform Fourier transform on the angle data using the Fourier angle decomposition method to decompose the angle components into the frequency domain, achieving high-precision measurement and noise suppression. Its algorithm formula is;

[0074] Among them, Θ(t) represents the angle data that changes with time or sampling points, Θ(ω) represents the representation of the angle data in the frequency domain, ω represents the angular frequency, and e -jωt represents the complex exponential kernel function;

[0075] Step 2.31 Use wavelet transform to extract multi-scale angle error features and perform hierarchical compensation to achieve precise control of angle pre-adjustment. Its algorithm formula is:

[0076] Among them, ψ(t) represents the mother wavelet function, f represents the scale factor, and g represents the translation factor;

[0077] By performing wavelet transform on the angle error signal θ e (t), the multi-resolution representation of the components θ e (t) at different scales is obtained:

[0078] Among them, represents the transformation coefficient obtained after performing a certain transformation on θ e (t), and j represents the translation in time;

[0079] Reconstruct the error signals at each scale through inverse wavelet transform:

[0080]

[0081] Step 2.32 Construct a real-time closed-loop control system, dynamically adjust the laser emission angle according to the angle error feedback, and combine fuzzy logic with adaptive control to quickly respond and correct the angle deviation, balancing the system response speed and robustness. Its algorithm formula is:

[0082]

[0083] Among them, u(t) represents the control output, e(t) represents the current angle error, represents the rate of change of the angle error, represents the membership function corresponding to the i-th fuzzy rule, describing the matching degree of the error and the rate of change under each rule, and μ i represents the control action corresponding to the i-th rule, and l represents the total number of fuzzy rules;

[0084] The control parameters adopt an adaptive update mechanism, and its formula is:

[0085]

[0086] Among them, η represents the adaptive learning rate, and K(t) represents the control gain at time t.

[0087] Preferably, the transmitting and receiving optical path module includes:

[0088] A coaxial optical path design unit, which is used to design and optimize the coaxial optical path for laser emission and reception, and ensure high fidelity and low diffraction loss of the laser beam during transmission. The specific steps include:

[0089] Step 3.1: Establish a numerical model of the coaxial optical path, simulate the propagation and diffraction effects of the light beam in each optical element, and realize fine light field reconstruction. The algorithm formula is:

[0090]

[0091] Among them, G n (x,y) represents the light field of the nth iteration, and respectively represent the Fourier transform and its inverse transform, H(u,v) represents the optical transfer function, and Ω(u,v) represents the target amplitude distribution;

[0092] Step 3.11: Use the interference principle to correct the phase differences of each optical element in the coaxial optical path, and make the wavefront of the synthesized light beam uniform. The algorithm formula is:

[0093]

[0094] Among them, φ n (x,y) represents the current wavefront phase, represents the difference between the target and the actual interference intensity distribution, and ∈ represents the step factor;

[0095] Step 3.12: Perform non-linear fitting through the optical path deviation data, adjust the positions of the optical elements, and realize the precise alignment of the coaxial optical path. The algorithm formula is:

[0096]

[0097] Among them, is the vector of optical parameters to be corrected, I meas represents the actually measured light intensity distribution, I model represents the light intensity distribution calculated by the model, I meas (x i ,y i ) represents the number of sampling points.

[0098] Preferably, the transmitting and receiving optical path module further includes:

[0099] The beam splitting prism unit is used to split and spectroscopically decompose the laser beam, and at the same time perform independent optical path correction and adjustment on different wavelength bands. The specific steps include:

[0100] Step 3.2 Use the beam splitting prism to decompose the laser beam according to wavelength, and calculate the refractive index n 2 (λ) using the Sellmeier equation, and determine the prism refraction angle. The algorithm formula is:

[0101]

[0102] where M represents the wavelength of light, m represents the number of terms of "sum" in the equation, represents the empirical fitting coefficient in the equation, O j represents the empirical fitting coefficient in the equation;

[0103] Step 3.21 Perform pre-adjustment control on the light beams of each wavelength band after beam splitting to compensate for the optical path error caused by dispersion. The algorithm formula is:

[0104] Δθ(M) = v1·(n(M) - n0) + v2,

[0105] where n0 represents the reference refractive index, and v1 and v2 represent the empirical correction coefficients;

[0106] Step 3.22 Reconstruct the light beams of each wavelength band into a unified output beam, and perform global optimization on the overall optical path. The algorithm formula is:

[0107]

[0108] where v represents the combined vector of each optical parameter, I combined represents the beam distribution after reconstruction, I target represents the target beam distribution;

[0109] The optical path calibration unit is responsible for precisely calibrating the entire transmitting and receiving optical path to ensure the long-term stable operation of the system and meet the high-precision requirements. The specific steps include:

[0110] Step 3.3 Use an interferometer or a high-resolution sensor to detect the deviation of each node in the optical path in real time;

[0111] Step 3.4 Perform data preprocessing on the detected deviation data, and use the robust regression algorithm for fitting and filtering. The algorithm formula is:

[0112]

[0113] where r represents the regression residual, and δ represents the threshold.

[0114] Preferably, the signal acquisition and processing module includes:

[0115] High-sensitivity detector unit: Use high-sensitivity detectors to capture raw data from the signal source; <o:p>< / o:p> <o:p>< / o:p>

[0116] High-speed data acquisition card unit: Used to quickly acquire signals, usually process high-speed signals, convert them into digital format and transmit them to the processing system; <o:p>< / o:p> <o:p>< / o:p>

[0117] Analog-to-digital conversion unit: Use an analog-to-digital converter to convert analog signals into digital signals; <o:p>< / o:p> <o:p>< / o:p>

[0118] Signal amplification and processing unit: Amplify, filter, and denoise the acquired signals. <o:p>< / o:p> <o:p>< / o:p>

[0119] The present invention provides a coaxial scanning ultra-long-distance detection lidar system. It has the following beneficial effects: <o:p>< / o:p> <o:p>< / o:p>

[0120] 1. By adopting a high-power 1550nm fiber laser, the present invention realizes ultra-long-distance detection on the premise of ensuring eye safety, and can achieve a detection range of more than 10km, thus breaking the limitation of the detection distance of traditional lidar, which is only 500m. <o:p>< / o:p> <o:p>< / o:p>

[0121] 2. By selecting a 1550nm laser that is safer for the human eye and significantly increasing the laser power to the kilowatt or even megawatt level, the present invention realizes a stronger echo signal, thereby significantly improving the detection signal-to-noise ratio. <o:p>< / o:p> <o:p>< / o:p>

[0122] 3. By using a microlens array to replace the traditional DOE diffractive optical element, and by flexibly adjusting the curvature radius and spacing of the microlenses, the present invention realizes efficient beam shaping of one-dimensional or two-dimensional lattices, while reducing the manufacturing cost and expanding the design freedom. <o:p>< / o:p> <o:p>< / o:p>

[0123] 4. By adopting a low field-of-view design and a scanning laser beam expander lens to achieve angle compression, the present invention effectively maintains the energy density of the transmitted beam, thereby meeting the requirements of ultra-long-distance high-precision detection. <o:p>< / o:p> <o:p>< / o:p>

[0124] 5. Through the coaxial optical path design, the present invention makes the transmission and reception share one optical path, and uses a beam splitting prism to separate the optical path. This not only reduces the system volume, but also reduces the cost of optical elements, and effectively avoids the problem of stray light interference caused by traditional dichroic mirrors. <o:p>< / o:p> <o:p>< / o:p>

[0125] 6. By integrating multiple advanced control algorithms such as fuzzy PID, adaptive control, genetic algorithm, Bayesian inference, and wavelet transform with a closed-loop feedback mechanism, the present invention realizes high-precision, real-time, and robust control of laser power regulation, stability control, beam shaping, and scanning angle regulation. <o:p>< / o:p> <o:p>< / o:p>

[0126] 7. Through the collaborative optimization of various modules such as the dynamic simulation of the laser energy level, the optical path propagation, the optical transmission, and the signal acquisition and processing in the present invention by using a combined objective function and multiple mathematical models, it is ensured that the overall system can still maintain high performance and stability in a complex environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0127] Figure 1 It is a flowchart of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0128] To enable those skilled in the art to understand the solution of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the scope of protection of the present invention.

[0129] The present invention will be described in detail below with reference to the accompanying drawings:

[0130] Embodiment:

[0131] Please refer to the appended Figure 1 , the embodiment of the present invention provides a coaxial scanning ultra-long-distance detection lidar system, including:

[0132] A laser light source module, which uses a 1550nm fiber laser to output high-power laser, ensuring sufficient laser energy and eye safety;

[0133] A beam shaping module, which converts the laser into a uniform and directional beam by using a microlens array according to the output laser, controls the divergence angle and maintains a high energy density;

[0134] A scanning beam expander lens module, which realizes long-distance coverage in a low field of view by dynamically adjusting the laser emission direction according to the directional beam;

[0135] A transmitting and receiving optical path module, which adopts a coaxial design according to the long-distance coverage, and separates the laser emission and the echo signal through a beam splitter prism;

[0136] A signal acquisition and processing module, which amplifies, acquires, and digitally processes the echo signal by means of a highly sensitive detector and a high-speed acquisition card according to the separated laser emission and echo signal, and uses an algorithm to realize the high-precision reconstruction of the distance and three-dimensional information;

[0137] S1, the laser light source module includes:

[0138] The 1550nm fiber laser unit is responsible for generating high-power laser output, ensuring sufficient laser energy, and is suitable for long-distance detection. In a fiber laser, the formula for the gain in the laser amplification process is:

[0139] G = exp[σ em N2L - σ abs N1L],

[0140] where σ em represents the emission cross-section, σ abs represents the absorption cross-section, N2 and N1 respectively represent the excited-state and ground-state atomic number densities, and L represents the effective length of the optical fiber;

[0141] Meanwhile, to accurately simulate the energy-level dynamics in the laser, the rate equations are used:

[0142]

[0143] where τ represents the laser lifetime, A eff represents the effective area, hν represents the photon energy, P pump represents the pump light power,, N total represents the total number density of doped ions in the laser medium, P signal represents the signal light power, represents the number of photons contained in the pump light per unit time, represents the rate at which ions in the excited state return to the ground state through spontaneous emission or non-radiative transitions, represents the number of photons in the signal light per unit time;

[0144] Laser power adjustment unit: The laser output power is detected by a high-speed photodetector. According to the error between the monitored value and the preset target, an adaptive fuzzy PID control algorithm is adopted to achieve the hybrid of traditional PID and fuzzy control. The formula form is:

[0145]

[0146] where u(t) represents the output control quantity, e(t) represents the power error, K p 、K i 、K d respectively represent the proportional, integral, and differential gains, α represents the weight coefficient, u fuzzy (t) is the control quantity output based on fuzzy rules, represents the time integral of the error, represents the rate of change of the error, u fuzzy (t) represents the output of the fuzzy controller, and (1 - α) represents the weighted coefficient complementary to α.

[0147] The laser light source module also includes:

[0148] The stability control unit collects the laser output fluctuation data in real time, conducts statistical and spectral analysis on the data, determines the fluctuation components, and realizes output stability by adjusting parameters such as pump current, temperature control, and fiber bending compensation. Therefore, a stability control strategy based on model predictive control is adopted, and its optimization problem expression is as follows:

[0149]

[0150] Among them, A represents the prediction horizon length, y(k) represents the predicted output, y ref (k) represents the reference output value, Q and R are the weight matrices of the state error and control input respectively, and u(k) represents the control variable. and represents the quadratic norm with the weight matrix, Q represents the output error weight matrix, and R represents the control input weight matrix. represents the measurement of the energy consumption of the control input at time k. represents the measurement of the deviation between the system output and the reference value at time k, and is weighted by the matrix Q. represents the accumulation of all errors and control consumptions from the current time to the next A - 1 steps.

[0151] [[ID=The human eye safety monitoring unit uses a high - speed photodetector to monitor the laser output power, compares the real - time detection value with the preset human eye safety threshold. When the output power exceeds the safety threshold, it immediately triggers the power automatic shutdown measure, and its algorithm formula is as follows:

[0152]

[0153] Among them, S(P) represents the logic function output, P represents the currently real - time detected laser output power, P safe represents the preset human eye safety threshold, and β represents the steepness adjustment factor; when P is lower than P safe , S(P) is close to 0, indicating low risk; when P exceeds P safe , S(P) quickly approaches 1, indicating a sharp rise in risk and triggering a safety protection response.

[0154] To prevent the system from repeatedly triggering or canceling protection due to small fluctuations in the critical interval, a hysteresis comparator can be introduced, and its logic formula is as follows:

[0155]

[0156] Among them, S t represents the final safety trigger status signal, θ upper and θ lowerRepresent the upper and lower thresholds of safety trigger and release trigger respectively, S prev Indicates the security status at the previous moment;

[0157] When the calculated risk coefficient S(P) exceeds the upper threshold θ upper When S(P) drops to the lower threshold θ lower When the value is below the threshold, the system will be released from the safe state. When the value is between the two thresholds, the previous state will be maintained to avoid frequent switching.

[0158] Specifically, by adopting a high-power 1550nm fiber laser, ultra-long-distance detection can be achieved while ensuring the safety of human eyes, with a detection range of more than 10km, thus breaking the limitation of traditional lidar's detection distance of only 500m. By selecting a 1550nm laser that is safer for the human eye and significantly increasing the laser power to kilowatts or even tens of thousands of watts, a stronger echo signal is achieved, thereby significantly improving the detection signal-to-noise ratio.

[0159] S2, beam shaping module includes:

[0160] The microlens array unit converts the original laser into a uniformly directional beam by precisely designing the arrangement and shape of the microlenses. It uses an algorithm based on Zernike polynomial expansion and minimum mean square error optimization to optimize the beam phase and maximize the uniformity of the beam, thereby providing good beam quality for long-distance transmission and high-precision detection. The algorithm formula is:

[0161]

[0162] Among them, I target (x, c) represents the intensity distribution of the target uniform spot, B(x, c) represents the amplitude distribution of the original laser beam, Z i (x,c) represents the i-th Zernike polynomial, which is used to represent the phase distortion of the light beam. i represents the Zernike coefficient to be optimized, represents the phase compensation parameter of each microlens, F represents the Fourier transform operator, F{·} represents an optical transmission operator, |·| represents the amplitude of the complex field, [·] 2 represents the quadratic loss, D represents the number of basis functions used for phase unwrapping;

[0163] The divergence angle control unit uses a hybrid optimization strategy of adaptive fuzzy control and genetic algorithm from beam parameter acquisition, error calculation, optical element adjustment to closed-loop feedback correction to achieve more precise divergence angle control in complex environments. The algorithm formula is:

[0164] Δθ=b(θ target -θ measured)+(1 - b)·GA(p),

[0165] where Δθ is the required divergence angle adjustment amount, θ target and θ measured respectively represent the target divergence angle and the actually measured divergence angle, b represents the weight between linear compensation and nonlinear genetic algorithm optimization, and GA(p) represents the nonlinear compensation term obtained by using the genetic algorithm (GA).

[0166] The beam shaping module further includes:

[0167] A beam uniformity adjustment unit that optimizes the beam energy distribution and improves the subsequent detection accuracy. The specific steps include:

[0168] Step 1.1 Use a high-resolution CCD to collect the intensity distribution of the beam cross-section;

[0169] Step 1.2 Divide the spot area into several sub-regions, calculate the energy proportion and overall mean in each region, and evaluate the uniformity of the beam energy distribution through Shannon entropy;

[0170] Step 1.3 Use a spatial light modulator to control the local phase and amplitude of the beam, and continuously update the control parameters according to the real-time feedback;

[0171] Step 1.4 Continuously compare the actually output beam with the preset target distribution through a closed-loop feedback mechanism, and iteratively optimize the adjustment parameters;

[0172] Step 1.5 Construct a joint objective function and iteratively optimize the control parameter vector u of the SLM:

[0173]

[0174] where p i (u) represents under the action of parameter u, p i represents the energy proportion in each region, the energy proportion of the i-th sub-region, M represents the total number of sub-regions, μ(I(u)) represents the beam energy mean under parameter u, μ target represents the target mean, λ represents the mean deviation compensation factor, ln(p i (u)) represents the natural logarithm of p i (u).

[0175] Specifically, by using a microlens array to replace the traditional DOE diffractive optical element, and utilizing the flexible control of the microlens curvature radius and spacing, efficient beam shaping of one-dimensional or two-dimensional lattices is achieved, while reducing the manufacturing cost and expanding the design freedom.

[0176] S3, The scanning beam expander lens module includes:

[0177] A afocal beam expander lens unit expands a laser beam in a non-focusing manner to cover a wider target area, while ensuring that the output light spot has a uniform energy distribution and low diffraction loss. The specific steps are as follows:

[0178] Step 2.1 Use a high-precision CCD to collect the original laser light field data, including amplitude and phase information;

[0179] Step 2.11 Determine the beam expansion ratio M according to the actual application requirements, and design the target output light field I target (x, y), requiring that the expanded light spot has a uniform energy distribution and low diffraction loss, while maintaining the overall energy stability;

[0180] Step 2.12 Use a spatial light modulator phase control element to apply a preset phase modulation to the incident light field to achieve afocal beam expansion;

[0181] Step 2.13 To simultaneously optimize the beam expansion matching effect and reduce the influence of diffraction and aberration, introduce a joint optimization objective function, and adjust the control parameter vector u to make the system reach the overall optimal state:

[0182]

[0183] where E in (x, y) represents the complex amplitude distribution of the original laser light field, e jφ(x,y,u) represents the phase control function controlled by the parameter vector u, represents the operator describing the propagation of the light field in the afocal beam expansion system, represents the ideal output intensity distribution obtained by scaling the coordinates of the target light field according to the beam expansion ratio M, represents the regularization term, and γ represents the regularization weight factor;

[0184] A dynamic scanning control unit adjusts the laser emission direction in real time to support long-distance detection in a low field of view. The specific steps are as follows:

[0185] Step 2.2 Use a sensor to collect the spatial distribution and propagation path data of the laser beam in real time, and update the beam state x k using Bayesian inference. The formula is:

[0186] where z k represents the real-time measurement data, P(z k ) represents the prior probability of observing z k , P(x k |x k-1 ) represents the state transition probability from the previous state x k-1 to the current state x k , and P(z k |xk ) represents the known state x k , in the case of which, the observation z k appears with a likelihood probability, P(x k |z k ) represents the posterior probability distribution of the system state x k after the given observation z k ;

[0187] Step 2.21 Calculate the relative position between the laser beam and the target area according to the prediction data, and quickly realize the preliminary direction correction through pre-adjustment. The algorithm formula is as follows:

[0188] wherein, represents the average value of the whole data, x i represents the estimated position of the i-th sampling, and E represents the number of samplings;

[0189] Step 2.22 Construct a closed-loop feedback system, monitor the deviation between the output beam and the target in real time, and dynamically adjust the scanning control parameters according to the error feedback. The algorithm formula is as follows:

[0190] wherein, e represents the direction error, F represents a positive definite matrix, and the stable regulation of the system is realized by minimizing V(e), represents the coefficient.

[0191] The scanning beam expander lens module further includes:

[0192] An angle control unit, which accurately controls the scanning angle to ensure that the detection range and resolution meet the requirements. The specific steps include:

[0193] Step 2.3 Collect the current laser emission angle data through a high-resolution angle sensor, and perform Fourier transform on the angle data by using the Fourier angle decomposition method to decompose the angle components into the frequency domain to achieve high-precision measurement and noise suppression. The algorithm formula is as follows;

[0194] wherein, Θ(t) represents the angle data that changes with time or sampling points, Θ(ω) represents the representation of the angle data in the frequency domain, ω represents the angular frequency, and e -jωt represents the complex exponential kernel function;

[0195] Step 2.31 Extract the multi-scale angle error features by using wavelet transform and perform hierarchical compensation to achieve precise control of angle pre-adjustment. The algorithm formula is as follows:

[0196] wherein, ψ(t) represents the mother wavelet function, f represents the scale factor, and g represents the translation factor;

[0197] By performing wavelet transform on the angular error signal θ e (t), multi-resolution representations of components θ e (t) at different scales are obtained:

[0198] Among them, represents the transform coefficient obtained after performing a certain transformation on θ e (t), and j represents the time translation;

[0199] The error signals at each scale are reconstructed through inverse wavelet transform:

[0200]

[0201] Step 2.32 Construct a real-time closed-loop control system. According to the angular error feedback, dynamically adjust the laser emission angle, and combine fuzzy logic and adaptive control to quickly respond to and correct the angular deviation, balancing the system response speed and robustness. Its arithmetic formula is:

[0202]

[0203] Among them, u(t) represents the control output, e(t) represents the current angular error, represents the rate of change of the angular error, represents the membership function corresponding to the i-th fuzzy rule, describing the matching degree of the error and the rate of change under each rule, μ i represents the control action corresponding to the i-th rule, and l represents the total number of fuzzy rules;

[0204] The control parameters adopt an adaptive update mechanism, and its formula is:

[0205]

[0206] Among them, η represents the adaptive learning rate, and K(t) represents the control gain at time t.

[0207] Specifically, by adopting a low field of view angle design and a scanning laser beam expander lens to achieve angular compression, the energy density of the emitted beam is effectively maintained, thus meeting the requirements of ultra-long distance and high-precision detection. By integrating multiple advanced control algorithms such as fuzzy PID, adaptive control, genetic algorithm, Bayesian inference, and wavelet transform with a closed-loop feedback mechanism, high-precision, real-time, and robust control of laser power adjustment, stability control, beam shaping, and scanning angle regulation is achieved.

[0208] S4, The transmitting and receiving optical path module includes:

[0209] Coaxial optical path design unit, which is used to design and optimize the coaxial optical path for laser emission and reception, ensuring high fidelity and low diffraction loss of the laser beam during transmission. The specific steps include:

[0210] Step 3.1: Establish a numerical model of the coaxial optical path, simulate the propagation and diffraction effects of the light beam in each optical element, and achieve fine light field reconstruction. The algorithm formula is:

[0211]

[0212] where G n (x,y) represents the light field of the nth iteration, and represent the Fourier transform and its inverse transform respectively, H(u,v) represents the optical transfer function, and Ω(u,v) represents the target amplitude distribution;

[0213] Step 3.11: Use the interference principle to correct the phase differences of each optical element in the coaxial optical path, making the wavefront of the synthesized light beam uniform. The algorithm formula is:

[0214]

[0215] where φ n (x,y) represents the current wavefront phase, represents the difference between the target and the actual interference intensity distribution, and ∈ represents the step factor;

[0216] Step 3.12: Perform non-linear fitting through the optical path deviation data, adjust the positions of the optical elements, and achieve precise alignment of the coaxial optical path. The algorithm formula is:

[0217]

[0218] where is the vector of optical parameters to be corrected, I meas represents the actually measured light intensity distribution, I model represents the light intensity distribution calculated by the model, and I meas (x i ,y i ) represents the number of sampling points.

[0219] The emission and reception optical path module also includes:

[0220] Beam splitting prism unit, which is used to split and spectroscopically decompose the laser beam, and at the same time perform independent optical path correction and adjustment for different wavelength bands. The specific steps include:

[0221] Step 3.2: Use the beam splitting prism to decompose the laser beam according to the wavelength, and calculate the refractive index n using the Sellmeier equation 2(λ), determine the prism refraction angle, and its algorithm formula is:

[0222]

[0223] where M represents the wavelength of light, m represents the number of terms of "sum" in the equation, represents the empirical fitting coefficient in the equation, O j represents the empirical fitting coefficient in the equation;

[0224] Step 3.21 performs pre-adjustment control on the light beams of each band after splitting, compensating for the optical path error caused by dispersion, and its algorithm formula is:

[0225] Δθ(M) = v1·(n(M) - n0) + v2,

[0226] where n0 represents the reference refractive index, and v1 and v2 represent the empirical correction coefficients;

[0227] Step 3.22 reconstructs the light beams of each band into a unified output light beam, and globally optimizes the overall optical path, and its algorithm formula is:

[0228]

[0229] where v represents the combined vector of each optical parameter, I combined represents the light beam distribution after reconstruction, I target represents the target light beam distribution;

[0230] The optical path calibration unit is responsible for precisely calibrating the entire transmitting and receiving optical path to ensure the long-term stable operation of the system and meet the high-precision requirements. The specific steps include:

[0231] Step 3.3 uses an interferometer or a high-resolution sensor to detect the deviation of each node in the optical path in real time;

[0232] Step 3.4 performs data preprocessing on the detected deviation data, and uses a robust regression algorithm for fitting and filtering, and its algorithm formula is:

[0233]

[0234] where r represents the regression residual, and δ represents the threshold.

[0235] Specifically, through the coaxial optical path design, the transmitting and receiving share the same optical path, and the optical path is separated by a beam splitting prism, which not only reduces the system volume, but also reduces the cost of optical components, and effectively avoids the problem of stray light interference caused by traditional dichroic mirrors.

[0236] S5, the signal acquisition and processing module includes:

[0237] High-sensitivity detector unit: A high-sensitivity detector is used to capture the original data from the signal source;

[0238] High-speed data acquisition card unit: It is used to quickly acquire signals, usually processes high-speed signals, converts them into digital format and transmits them to the processing system;

[0239] Analog-to-digital conversion unit: An analog-to-digital converter is used to convert analog signals into digital signals;

[0240] Signal amplification and processing unit: Processes the acquired signals by amplifying, filtering, and denoising them. Although the embodiments of the present invention have been shown and described, for those of ordinary skill in the art, it can be understood that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A coaxial scanning ultra-long-distance detection lidar system, characterized in that, Including: The laser light source module uses a 1550nm fiber laser to output high-power laser, ensuring sufficient laser energy and eye safety; The beam shaping module converts the laser into a uniform and directional beam using a microlens array according to the output laser, controls the divergence angle and maintains a high energy density; The scanning beam expander lens module realizes long-distance coverage in a low field of view by dynamically adjusting the laser output direction according to the directional beam; The transmitting and receiving optical path module adopts a coaxial design according to the long-distance coverage, and separates the laser emission and echo signals through a beam splitter prism; The signal acquisition and processing module amplifies, acquires and digitally processes the echo signal by means of a high-sensitivity detector and a high-speed acquisition card according to the separated laser emission and echo signals, and uses an algorithm to achieve high-precision reconstruction of distance and three-dimensional information.

2. The coaxial scanning ultra-long distance detection lidar system according to claim 1, characterized in that The laser light source module includes: A 1550nm fiber laser unit responsible for generating high-power laser output, ensuring sufficient laser energy, suitable for long-distance detection. In the fiber laser, the calculation formula for the gain in the laser amplification process is: G = exp[σ em N2L - σ abs N1L], Among them, σ em represents the emission cross-section, and σ abs represents the absorption cross-section. N2 and N1 respectively represent the excited state and ground state atomic number densities, and L represents the effective length of the optical fiber; At the same time, to accurately simulate the energy level dynamics in the laser, the rate equation is used: Among them, τ represents the laser lifetime, A eff represents the effective area, hν represents the photon energy, P pump represents the pump light power, and N total represents the total number density of doped ions in the laser medium, P signal represents the signal light power, represents the number of photons contained in the pump light per unit time, represents the rate at which ions in the excited state return to the ground state by spontaneous emission or through non-radiative transitions, represents the number of photons in the signal light per unit time; Laser power adjustment unit: Detect the laser output power through a high-speed photodetector, and adopt an adaptive fuzzy PID control algorithm according to the error between the monitored value and the preset target to realize the hybrid of traditional PID and fuzzy control. The formula form is: where, u(t) represents the output control quantity, e(t) represents the power error, K p , K i , K d represent the proportional, integral, and derivative gains respectively, α represents the weight coefficient, u fuzzy (t) is the control quantity output based on the fuzzy rules, represents the time integral of the error, represents the rate of change of the error, u fuzzy (t) represents the output of the fuzzy controller, and (1 - α) represents the weighted coefficient complementary to α.

3. The coaxial scanning ultra-long distance detection lidar system according to claim 1, characterized in that The laser light source module also includes: The stability control unit collects the laser output fluctuation data in real time, statistically analyzes and spectroscopically analyzes the data to determine the fluctuation components. According to the fluctuation characteristics, the output stability is achieved by adjusting parameters such as pump current, temperature control and fiber bending compensation. Therefore, a stability control strategy based on model predictive control is adopted, and its optimization problem expression is: where A represents the prediction horizon length, y(k) represents the predicted output, and y ref (k) represents the reference output value. Q and R are the weight matrices of the state error and the control input respectively, and u(k) represents the control variable. and represents the quadratic norm with a weighting matrix, Q represents the output error weighting matrix, and R represents the control input weighting matrix. represents the measurement of the energy consumption of the control input at time k. represents the measurement of the deviation between the system output and the reference value at time k, and is weighted by the matrix Q. represents the accumulation of all errors and control consumptions from the current time to A-1 steps in the future. The eye safety monitoring unit monitors the laser output power using a high-speed photodetector, compares the real-time detected value with the preset eye safety threshold, and immediately triggers the power automatic shutdown measure when the output power exceeds the safety threshold. Its algorithm formula is: Among them, S(P) represents the output of the logic function, P represents the currently detected real-time laser output power, and P safe represents the preset eye safety threshold, and β represents the steepness adjustment factor; when P is lower than P safe , S(P) is close to 0, indicating low risk; when P exceeds P safe , S(P) rapidly approaches 1, indicating a sharp increase in risk and triggering a safety protection response; To prevent the system from repeatedly triggering or canceling protection due to small fluctuations in the critical interval, a hysteresis comparator can be introduced, and its logic formula is: Among them, S t represents the final safety trigger status signal, θ upper and θ lower respectively represent the upper and lower thresholds for safety triggering and de - triggering, and S prev represents the safety status at the previous moment; When the calculated risk coefficient S(P) exceeds the upper threshold θ upper the system enters the safe state. When S(P) drops below the lower threshold θ lower the system exits the safe state. When it is between the two thresholds, it maintains the previous state to avoid frequent switching.

4. The coaxial scanning ultra-long-distance detection lidar system according to claim 1, wherein, The beam shaping module includes: The microlens array unit realizes the conversion of the original laser into a uniform and directional beam by precisely designing the arrangement and shape of the microlenses. An algorithm based on Zernike polynomial expansion and least mean square error optimization is used to optimize the beam phase and maximize the beam uniformity, thus providing good beam quality for long-distance transmission and high-precision detection. Its algorithm formula is: Among them, I target (x, c) represents the intensity distribution of the target uniform light spot, B(x, c) represents the amplitude distribution of the original laser beam, Z i (x, c) represents the i-th Zernike polynomial, which is used to represent the phase distortion of the beam, a i represents the Zernike coefficient to be optimized, representing the phase compensation parameter of each microlens, F represents the Fourier transform operator, F{·} represents a certain optical transmission operator, |·| represents taking the amplitude of the complex field, [·] 2 represents the quadratic loss, and D represents the number of basis functions used to expand the phase; The divergence angle control unit adopts a hybrid optimization strategy of adaptive fuzzy control and genetic algorithm from beam parameter acquisition, error calculation, optical element adjustment to closed-loop feedback correction to achieve more refined divergence angle control in a complex environment. Its algorithm formula is: Δθ = b(θ target - θ measured ) + (1 - b)·GA(p), Among them, Δθ is the required divergence angle adjustment amount, θ target and θ measured respectively represent the target divergence angle and the actually measured divergence angle, b represents the weight between linear compensation and nonlinear genetic algorithm optimization, and GA(p) represents the nonlinear compensation term obtained by using the genetic algorithm (GA).

5. A coaxial scanning ultra-long distance detection lidar system according to claim 1, wherein The beam shaping module also includes: The beam uniformity adjustment unit optimizes the beam energy distribution and improves the subsequent detection accuracy. The specific steps include: Step 1.1 Use a high-resolution CCD to collect the intensity distribution of the beam cross-section; Step 1.2 Divide the light spot area into several sub - areas, calculate the energy proportion and overall mean value in each area, and evaluate the uniformity of the beam energy distribution through Shannon entropy; Step 1.3 Use a spatial light modulator to regulate the local phase and amplitude of the beam, and continuously update the control parameters according to real - time feedback; Step 1.4 Continuously compare the actual output beam with the preset target distribution through a closed - loop feedback mechanism, and iteratively optimize the adjustment parameters; Step 1.5 Construct a joint objective function, and iteratively optimize the control parameter vector u of the SLM: Among them, p i (u) represents p under the action of parameter u i represents the energy proportion in each region, the energy proportion of the i-th sub-region, M represents the total number of sub-regions, μ(I(u)) represents the mean value of the beam energy under parameter u, μ target represents the target mean value, λ represents the mean deviation compensation factor, ln(p i (u)) represents the natural logarithm of p i (u).

6. The coaxial scanning ultra-long distance detection lidar system according to claim 1, wherein The scanning beam expander lens module includes: A non - focal beam expander lens unit that non - focally expands the laser beam to cover a wider target area, while ensuring that the output light spot has a uniform energy distribution and low diffraction loss. The specific steps are as follows: Step 2.1 Use a high - precision CCD to collect the original laser light field data, including amplitude and phase information; Step 2.11 Determine the beam expansion ratio M according to the actual application requirements, and design the target output optical field I target (x, y), requiring that the expanded spot has a uniform energy distribution and low diffraction loss, while maintaining the overall energy stability; Step 2.12 Use a spatial light modulator phase control element to apply a preset phase modulation to the incident light field to achieve non - focal beam expansion; Step 2.13 To simultaneously optimize the beam expansion matching effect and reduce the influence of diffraction and aberration, introduce a joint optimization objective function, and adjust the control parameter vector u to make the system reach an overall optimal state: Among them, E in (x, y) represents the complex amplitude distribution of the original laser light field, and e jφ(x,y,u) represents the phase modulation function controlled by the parameter vector u, represents the operator describing the propagation of the light field in the afocal beam expander system, represents the ideal output intensity distribution after coordinate scaling of the target light field according to the beam expansion ratio M, represents the regularization term, and γ represents the regularization weight factor; A dynamic scanning control unit that adjusts the laser emission direction in real time to support long - distance detection under a low field of view. The specific steps are as follows: Step 2.2 Use sensors to collect the spatial distribution and propagation path data of the laser beam in real time, and use Bayesian inference to update the beam state x k as follows: Among them, z k represents real-time measurement data, and P(z k ) represents the prior probability of observing z k . P(x k |x k-1 ) represents the state transition probability of transitioning from the state x k-1 at the previous moment to the state x k at this moment. P(z k |x k ) represents the likelihood probability of observing z k when the state x k is known. P(x k |z k ) represents the posterior probability distribution of the system state x k after the given observation z k ; Step 2.21 Calculate the relative position between the laser beam and the target area based on the prediction data, and quickly achieve preliminary direction correction through pre-adjustment. The algorithm formula is as follows: Among them, represents the average value of the entire data, and x i represents the estimated position of the i-th sampling, and E represents the number of samplings; Step 2.22 Construct a closed-loop feedback system to monitor the deviation between the output beam and the target in real time, and dynamically adjust the scanning control parameters according to the error feedback. Its algorithm formula is as follows: Among them, e represents the direction error, F represents a positive definite matrix, and the stable regulation of the system is achieved by minimizing V(e). represents the coefficient.

7. The coaxial scanning type ultra-long distance detection lidar system according to claim 1, wherein, The scanning beam expander lens module also includes: The angle control unit that precisely controls the scanning angle to ensure that the detection range and resolution meet the requirements. The specific steps include: Step 2.3 Collect the current laser emission angle data through a high-resolution angle sensor, and perform Fourier transform on the angle data using the Fourier angle decomposition method to decompose the angle components into the frequency domain, achieving high-precision measurement and noise suppression. Its algorithm formula is; Among them, Θ(t) represents the angular data that changes with time or sampling points, Θ(ω) represents the representation of the angular data in the frequency domain, ω represents the angular frequency, and e -jωt represents the complex exponential kernel function; Step 2.31 uses wavelet transform to extract multi-scale angular error features and performs hierarchical compensation to achieve precise control of angle pre-adjustment. Its algorithm formula is as follows: Among them, ψ(t) represents the mother wavelet function, f represents the scale factor, and g represents the translation factor; By performing wavelet transform on the angular error signal θ e (t), the multi-resolution representation of the components θ e (t) at different scales is obtained: Among them, represents the transformation coefficient obtained after making a certain change to θ e (t), and j represents the time shift; Reconstruct the error signals of each scale through inverse wavelet transform: Step 2.32 Construct a real - time closed - loop control system, dynamically adjust the laser emission angle according to the angle error feedback, and combine fuzzy logic with adaptive control to quickly respond and correct the angle deviation, balancing the system response speed and robustness. Its algorithm formula is: where \(u(t)\) represents the control output and \(e(t)\) represents the current angular error. represents the rate of change of the angular error. represents the membership function corresponding to the \(i\)-th fuzzy rule, which describes the matching degree of the error and the rate of change under each rule, \(\mu\) i represents the control action corresponding to the \(i\)-th rule, and \(l\) represents the total number of fuzzy rules. The control parameters adopt an adaptive update mechanism, and its formula is: Among them, η represents the adaptive learning rate, and K(t) represents the control gain at time t.

8. A coaxial scanning ultra-long distance detection lidar system according to claim 1, characterized in that, The transmitting and receiving optical path module includes: A coaxial optical path design unit used to design and optimize the coaxial optical path for laser emission and reception, ensuring high - fidelity and low diffraction loss of the laser beam during transmission. The specific steps include: Step 3.1 Establish a numerical model of the coaxial optical path, simulate the propagation and diffraction effects of the beam in each optical element, and achieve fine light field reconstruction. Its algorithm formula is: Among them, G n (x, y) represents the optical field at the nth iteration, θ and θ -1 respectively represent the Fourier transform and its inverse transform, H(u, v) represents the optical transfer function, and Ω(u, v) represents the target amplitude distribution; Step 3.11 Use the interference principle to correct the phase differences of each optical element in the coaxial optical path to make the wavefront of the synthesized beam uniform. Its algorithm formula is: where φ n (x, y) represents the current wavefront phase, represents the difference between the target and the actual interference intensity distribution, and ∈ represents the step factor; Step 3.12 Perform non - linear fitting on the optical path deviation data to adjust the positions of the optical elements to achieve precise alignment of the coaxial optical path. Its algorithm formula is: Among them, is the optical parameter vector to be corrected, I meas represents the actually measured light intensity distribution, I model represents the light intensity distribution calculated by the model, I meas (x i , y i ) represents the number of sampling points.

9. The coaxial scanning ultra-long-distance detection lidar system according to claim 1, characterized in that, The transmitting and receiving optical path module also includes: The beam - splitting prism unit used to split and spectroscopically decompose the laser beam, and at the same time perform independent optical path correction and adjustment on different bands. The specific steps include: Step 3.2 Use a beam splitter prism to decompose the laser beam by wavelength, and use the Sellmeier equation to calculate the refractive index n 2 (λ), and determine the prism refraction angle. The algorithm formula is as follows: Among them, M represents the wavelength of light, m represents the number of terms of "sum" in the equation, represents the empirical fitting coefficient in the equation, O j represents the empirical fitting coefficient in the equation; Step 3.21 Perform pre - adjustment control on the beam of each band after beam splitting to compensate for the optical path error caused by dispersion. Its algorithm formula is: Δθ(M) = v1·(n(M) - n0) + v2, where n0 represents the reference refractive index, and v1 and v2 represent empirical correction coefficients; Step 3.22 reconstructs the light beams of each band into a unified output light beam and globally optimizes the overall optical path. The algorithm formula is as follows: where \(v\) represents the combined vector of each optical parameter, \(I\) combined represents the reconstructed beam distribution, and \(I\) target represents the target beam distribution; The optical path calibration unit is responsible for precisely calibrating the entire transmitting and receiving optical path to ensure the long-term stable operation of the system and meet the high-precision requirements. The specific steps include: Step 3.3 uses an interferometer or a high-resolution sensor to detect the deviations of each node in the optical path in real time; Step 3.4 performs data preprocessing on the detected deviation data and uses a robust regression algorithm for fitting and filtering. The algorithm formula is as follows: where r represents the regression residual and δ represents the threshold.

10. The coaxial scanning ultra-long distance detection lidar system according to claim 1, characterized in that, The signal acquisition and processing module includes: High-sensitivity detector unit: uses a high-sensitivity detector to capture the original data from the signal source; High-speed data acquisition card unit: used to quickly acquire signals, usually processes high-speed signals, converts them into digital formats and transmits them to the processing system; Analog-to-digital conversion unit: uses an analog-to-digital converter to convert analog signals into digital signals; Signal amplification and processing unit: amplifies, filters, and denoises the acquired signals.

Images