Spot-adjustable laser remote sensing optical system

Abstract

The application discloses a laser remote sensing optical machine system with adjustable light spots and relates to the technical field of light spot adjustment. The laser remote sensing optical machine system with adjustable light spots comprises a laser emission unit for emitting a laser beam, a detection optical lens adjusting unit for receiving the laser beam emitted by the laser emission unit and performing shape control, a receiving lens for receiving the light signal returned by the laser beam after reaching a target after the shape control of the detection optical lens adjusting unit and performing focusing, a detection unit for converting the light signal focused by the receiving lens into an electric signal, and a signal processing unit for receiving the electric signal output by the detection unit and analyzing and generating a regulation and control instruction. The laser emission unit, the detection optical lens adjusting unit and the micro-displacement adjusting platform are intelligently regulated and controlled by the signal processing unit, so that the automatic adjustment of the focal length is realized, the size of the light spot is controlled, the detection requirements of different distances are adapted, and the optical machine signal quality is automatically adjusted to be optimal.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of light spot adjustment, in particular to a light spot adjustable laser remote sensing optical system. BACKGROUND

[0002] Most of the remote sensing optical systems on the market are traditional optical systems, usually designed with fixed focal length, and the light spot is difficult to adjust to adapt to its flexibility and application range in long-distance and short-distance detection. Especially when it is necessary to cover a large range of detection area and perform high-precision detection at the same time, the limitations of the prior art are particularly obvious. Moreover, most systems are equipped with only a single lens, which makes it difficult to switch the beam shape, limiting the detection range and flexibility. At the same time, during the debugging of the optical system, the position of the detector needs to be adjusted manually, which is low in efficiency and difficult to ensure signal stability. Therefore, there is an urgent need for a remote sensing optical system that can automatically adjust the focal length, switch the beam type and ensure the quality of the optical signal to be automatically adjusted to the best.

[0003] The limitations of the prior art at least include the following problems. The position of the photoelectric detector is usually a fixed structure, which is difficult to adjust in real time according to the changes in the optical path or the signal state. This structure is particularly prominent in a variable environment. For example, when performing medium and long distance remote sensing, the light spot focal point is easily shifted. If the detector is still located at a static fixed position, it will directly lead to a significant decrease in the received optical signal intensity, a poor signal-to-noise ratio, and difficulty in capturing effective signals. Especially in the case of uneven reflectivity of the target surface or slight fluctuation of the laser divergence angle, the signal focal point and the receiving surface are more likely to be separated. The system is difficult to follow the optimal landing point of the signal, causing the error to soar, thereby making the measurement result lack of accuracy. Moreover, the optical system posture often needs to be manually adjusted, which is not only low in efficiency, but also difficult to realize remote and dynamic application, seriously restricting the practicality and adaptability of high-precision remote sensing systems in complex working conditions. SUMMARY

[0004] In view of the deficiencies of the prior art, the present application provides a light spot adjustable laser remote sensing optical system, which solves the problem that the photoelectric detector is difficult to dynamically adjust the position in the prior art, thereby causing insufficient remote sensing accuracy and adaptability.

[0005] To achieve the above purpose, the present application realizes the following technical scheme: a light spot adjustable laser remote sensing optical system, comprising: a laser emitting unit for emitting a laser beam; a detection optical lens adjusting unit for receiving the laser beam emitted by the laser emitting unit and controlling the shape; a receiving lens for receiving the light signal returned by the laser beam after reaching the target after being controlled in shape by the detection optical lens adjusting unit and focusing; a detection unit for converting the light signal focused by the receiving lens into an electrical signal; and a signal processing unit for receiving the electrical signal output by the detection unit and analyzing and generating a control instruction.

[0006] Further, the detection unit includes a photodetector receiving the focused light signal of the receiving lens, and is stably carried by a detector mounting unit, the photodetector and the detector mounting unit are positionally adjusted by a micro-displacement adjustment platform.

[0007] Further, the detection light lens adjustment unit includes a one-letter lens for adjusting the shape of the laser beam emitted by the laser emission unit, and an aspherical collimating lens, the one-letter lens and the aspherical collimating lens are switched by a lens adjustment frame.

[0008] Further, the signal processing unit includes: a data acquisition module for acquiring state data of the laser remote sensing optical machine to be adjusted in real time, the state data including electrical signal data and component position data set; a data analysis module for respectively performing feature analysis on the state data of the laser remote sensing optical machine to be adjusted, obtaining a remote sensing evaluation index set of the laser remote sensing optical machine to be adjusted, including a remote sensing response accuracy index and a structure optical matching index, and performing comprehensive analysis to obtain a remote sensing credibility index of the laser remote sensing optical machine to be adjusted; and an optical machine adjustment module for intelligently regulating and controlling the displacement adjustment unit, the detection light lens adjustment unit and the micro-displacement adjustment platform of the laser remote sensing optical machine to be adjusted based on the remote sensing credibility index.

[0009] Further, the electrical signal data is specifically the signal amplitude at each time point, and the specific steps of obtaining the remote sensing response accuracy index of the laser remote sensing optical machine to be adjusted are as follows: obtaining the detection three-dimensional position coordinates of the laser remote sensing optical machine to be adjusted, and inputting the electrical signal data and the detection three-dimensional position coordinates into a pre-trained position optimization model for optimization analysis to obtain the detection optimized three-dimensional position coordinates of the laser remote sensing optical machine to be adjusted, and combining the detection three-dimensional position coordinates for comprehensive analysis to obtain a detection position deviation index of the laser remote sensing optical machine to be adjusted; based on the electrical signal data of the laser remote sensing optical machine to be adjusted, analyzing the signal response index of the laser remote sensing optical machine to be adjusted, and combining the position deviation index to analyze the remote sensing response accuracy index of the laser remote sensing optical machine to be adjusted.

[0010] Further, the position optimization model is specifically a convolutional neural network, including an input layer, a convolutional layer, a fully connected embedding layer, a fully connected fusion layer, and an output prediction layer. The specific steps for obtaining the optimized three-dimensional position coordinates of the detection of the laser remote sensing optical machine to be adjusted are as follows: in the input layer of the convolutional neural network, the three-dimensional position coordinates of the detection of the laser remote sensing optical machine to be adjusted and the electrical signal data are received and preprocessed; in the convolutional layer of the convolutional neural network, one-dimensional convolution processing is performed on the preprocessed electrical signal data of the laser remote sensing optical machine to be adjusted, to obtain a detection signal feature vector of the laser remote sensing optical machine to be adjusted; in the fully connected embedding layer of the convolutional neural network, nonlinear mapping processing is performed on the preprocessed three-dimensional position coordinates of the detection of the laser remote sensing optical machine to be adjusted, to obtain a detection space embedding vector of the laser remote sensing optical machine to be adjusted; in the fully connected fusion layer of the convolutional neural network, the detection feature vector and the detection space embedding vector of the laser remote sensing optical machine to be adjusted are fused, to obtain a detection comprehensive state feature vector of the laser remote sensing optical machine to be adjusted; and in the output prediction layer of the convolutional neural network, prediction processing is performed on the detection comprehensive state feature vector of the laser remote sensing optical machine to be adjusted, to obtain the optimized three-dimensional position coordinates of the detection of the laser remote sensing optical machine to be adjusted.

[0011] Further, the specific formula for calculating the remote sensing response accuracy index of the laser remote sensing optical machine to be adjusted is as follows: ; wherein, is the remote sensing response accuracy index of the laser remote sensing optical machine to be adjusted, is the signal response index of the laser remote sensing optical machine to be adjusted, is the response adjustment coefficient stored in the database, is the position deviation index of the laser remote sensing optical machine to be adjusted, is the position deviation adjustment coefficient stored in the database, is the interaction adjustment coefficient stored in the database.

[0012] Further, the component position data set includes the three-dimensional position coordinates of the laser, the one-letter lens, the aspheric collimating lens, the receiving lens, the target region boundary point, and the target center point. The specific steps for obtaining the structural optical matching index of the laser remote sensing optical machine to be adjusted are as follows: based on the component position data set of the laser remote sensing optical machine to be adjusted, the spot shape adaptation index and the optical path alignment index of the laser remote sensing optical machine to be adjusted are analyzed; and the spot shape adaptation index and the optical path alignment index of the laser remote sensing optical machine to be adjusted are analyzed by weighting, to obtain the structural optical matching index of the adjusted remote sensing optical machine.

[0013] Further, the specific formula for calculating the remote sensing confidence index of the laser remote sensing optical machine to be adjusted is as follows: ; wherein, is the remote sensing confidence index of the laser remote sensing optical machine to be adjusted, A remote sensing response precision index of a laser remote sensing optical machine to be adjusted, A signal quality adjustment coefficient stored in the database, A structure optical matching index of a laser remote sensing optical machine to be adjusted, An optical matching adjustment coefficient stored in the database, A difference adjustment coefficient stored in the database, A superposition adjustment coefficient stored in the database.

[0014] Further, the specific steps of intelligently regulating the displacement adjustment unit, the detection optical lens adjustment unit and the micro-displacement adjustment platform of the laser remote sensing optical machine to be adjusted based on the remote sensing reliability index are as follows: the remote sensing reliability index of the laser remote sensing optical machine to be adjusted is judged and analyzed with the preset remote sensing reliability index threshold value; if the remote sensing reliability index of the laser remote sensing optical machine to be adjusted is higher than the preset remote sensing reliability index threshold value, no regulation is performed; if the remote sensing reliability index of the laser remote sensing optical machine to be adjusted is lower than or equal to the preset remote sensing reliability index threshold value, the displacement adjustment unit, the detection optical lens adjustment unit and the micro-displacement adjustment platform of the laser remote sensing optical machine to be adjusted are regulated and processed.

[0015] The present application has the following advantages:

[0016] (1) The laser remote sensing optical machine system with adjustable light spot, by setting the micro-displacement adjustment platform, realizes high-precision micro-position adjustment of the photodetector in three-dimensional space, based on the current electrical signal amplitude sequence of the photodetector and the spatial coordinate input of the micro-displacement adjustment platform, analyzes and outputs the detection position with the optimal signal response in the current state, after the micro-displacement adjustment platform receives the optimal position coordinate instruction, drives the detector to realize synchronous three-axis micro-shift, so that it quickly aligns with the point with the strongest signal, improves the signal amplitude and stability, thereby avoiding the light deviation caused by environmental disturbance or target displacement, and the system adjustment process has fast response speed, high precision and stable operation, to ensure that the photodetector always keeps aligned with the focal point during dynamic measurement, thereby enhancing the reliability and continuity of the remote sensing signal, and effectively supporting the quality requirements of subsequent data analysis.

[0017] (2) The light spot adjustable laser remote sensing light machine system, by constructing a controllable light spot shape switching mechanism, by setting a rotating motor to drive a lens adjustment frame, realizing the automatic switching of aspheric lens and linear lens, and based on the analysis of the structure optical matching index obtained from the component position data set, determining the automatic switching of the lens, so as to switch different light spot structures as needed, and then making the system not only be used for high-precision focal point measurement, but also be used for wide scanning of planar or linear targets, thereby having strong form adaptability, and the light spot output is flexible and fast in response, which is convenient for unified remote sensing processing of multi-scale and multi-structure targets, so as to improve the application efficiency and measurement coverage capability of the system in the scene of complex structure detection of space components and multi-region dynamic analysis.

[0018] (3) The light spot adjustable laser remote sensing light machine system, by constructing an intelligent judgment mechanism based on a remote sensing credibility index, fusing the remote sensing response accuracy index and the structure optical matching index for evaluation, forming a comprehensive state decision parameter, and taking the remote sensing credibility index as a trigger condition to guide whether the system starts the regulation process, if the index is lower than the set threshold, the system will automatically control the axial position of the laser, the state of the light spot lens and the position of the detector, and the regulation of each link is based on the current signal improvement rate for feedback judgment to ensure that the adjustment has clear signal improvement, and a complete control closed loop from state recognition, decision triggering, parameter adjustment to signal response verification is constructed, so that the system has full-process adaptive ability, thereby having the characteristics of rapid regulation and continuous optimization in complex remote sensing scenes, and further improving the intelligent level of the whole system.

[0019] Of course, it is not necessary for any product implementing the present application to achieve all the advantages mentioned above. BRIEF DESCRIPTION OF DRAWINGS

[0020] Figure 1 It is a schematic diagram of the principle of the light spot adjustable laser remote sensing light machine system of the present application.

[0021] Figure 2 It is a schematic diagram of the structure of the detection unit of the present application.

[0022] Figure 3 It is a schematic diagram of the structure of the laser emission unit of the present application.

[0023] Figure 4 It is a schematic diagram of the structure of the detection light lens adjustment unit of the present application.

[0024] Figure 5 It is a schematic diagram of the light spot switching of the present application.

[0025] Figure 6 It is a specific step flowchart of obtaining the remote sensing response accuracy index of the laser remote sensing light machine to be adjusted in the light spot adjustable laser remote sensing light machine system of the present application.

[0026] Figure 7 A specific step flow chart for obtaining the structural optical matching index of the laser remote sensing optical machine to be adjusted in the spot adjustable laser remote sensing optical machine system of the present application.

[0027] In the figure, 1, laser emission unit; 101, laser; 102, displacement adjustment unit; 2, detection optical lens adjustment unit; 201, one-letter lens; 202, lens adjustment frame; 203, aspheric collimating lens; 204, rotary motor; 3, receiving lens; 4, indicating light lens adjustment unit; 5, indicating red light; 6, signal processing unit; 7, detection unit; 701, photodetector; 702, detector mounting unit; 703, micro-displacement adjustment platform. DETAILED DESCRIPTION

[0028] Please refer to Figures 1-5 The embodiment of the present application provides a technical scheme: a spot adjustable laser remote sensing optical machine system, comprising: a laser emission unit 1 for emitting a laser beam with adjustable focusing characteristics, the laser emission unit 1 comprising a laser 101 for emitting a laser beam, the laser 101 performing precise displacement along the laser optical axis direction based on a displacement adjustment unit 102, thereby changing the diameter and divergence angle of the laser output spot to realize precise control of the spot size; a detection optical lens adjustment unit 2 for receiving the laser beam emitted by the laser emission unit 1 and performing shape control; a receiving lens 3 for receiving the light signal returned by the laser beam after reaching the target after being shape-controlled by the detection optical lens adjustment unit 2, and focusing; an indicating red light 5 for providing a visible light beam consistent with the output path of the laser emission unit 1; an indicating light lens adjustment unit 4 for adjusting the exit direction of the indicating red light 5 to make it consistent with the path of the laser beam after being acted on by the detection optical lens adjustment unit 2, thereby visualizing the invisible laser landing point; a detection unit 7 for converting the light signal focused by the receiving lens 3 into an electrical signal; a signal processing unit 6 for receiving the electrical signal output by the detection unit 7 and analyzing to generate control instructions for intelligent control of the laser emission unit 1 and the detection optical lens adjustment unit 2.

[0029] The detection unit 7 comprises a photodetector 701 receiving the focused light signal of the receiving lens 3, and is stably carried by a detector mounting unit 702 to ensure the heat dissipation and working stability of the photodetector 701, the photodetector 701 and the detector mounting unit 702 are adjusted in a small position by a micro-displacement adjustment platform 703, so that the signal state of the photodetector 701 reaches the optimum.

[0030] The probe light lens adjusting unit 2 comprises a one-word lens 201 for adjusting the shape of the laser beam emitted by the laser emitting unit 1, and an aspheric collimating lens 203, the one-word lens 201 and the aspheric collimating lens 203 are switched by a lens adjusting frame 202, the lens adjusting frame 202 is driven by a rotating motor 204, the rotating motor 204 drives the lens adjusting frame 202 to rotate, so as to realize the switching of the one-word lens 201 and the aspheric collimating lens 203 in the optical path, thereby completing the switching output of different spot shapes.

[0031] Specifically, the signal processing unit 6 comprises: a data acquisition module, configured to acquire state data of the laser remote sensing optical system to be adjusted in real time (i.e. a real-time feedback data set acquired in each laser emission to receiving behavior), the state data comprising electrical signal data (converted by the photodetector 701), a component position data set; a data analysis module, configured to perform feature analysis on the state data of the laser remote sensing optical system to be adjusted respectively, to obtain a remote sensing evaluation index set of the laser remote sensing optical system to be adjusted, comprising a remote sensing response accuracy index, a structure optical matching index, and comprehensively analyzing to obtain a remote sensing credibility index of the laser remote sensing optical system to be adjusted; and an optical system adjusting module, configured to intelligently control the laser emitting unit 1 (the displacement adjusting unit 102 therein), the probe light lens adjusting unit 2, and the micro-displacement adjusting platform 703 of the laser remote sensing optical system to be adjusted based on the remote sensing credibility index.

[0032] In the embodiment, the electrical signal data is acquired by being converted in real time by the photodetector 701, the photodetector receives the optical signal focused by the receiving lens 3, and converts the optical signal into corresponding electrical signal, the component position data set, i.e. the position coordinates of each key component of the remote sensing optical system, specifically, the three-dimensional position coordinates of each component such as the photodetector 701, the laser emitting unit 102, and the probe light lens adjusting unit 2 are collected in real time by the built-in positioning sensor or external tracking device, then the data analysis module in the signal processing unit 6 further analyzes these data in depth to generate the remote sensing response accuracy index, the structure optical matching index, and the remote sensing credibility index, so that the system can evaluate the signal quality and optical matching in real time, and make corresponding intelligent adjustment decision, thereby ensuring the optimal signal reception and measurement accuracy in the remote sensing process, finally, the optical system adjusting module accurately controls the laser emitting unit 102, the probe light lens adjusting unit 2, and the micro-displacement adjusting platform 703 according to the calculated remote sensing credibility index, and intelligently adjusts the positions and states of each component, so that the system can optimize the optical system setting in real time under dynamic environment, thereby avoiding the influence of spot deviation, unstable signal and other problems on the measurement result, and improving the self-adaptive ability of the remote sensing optical system under complex conditions, thereby enabling it to maintain high precision and high stability in the variable measurement scene.

[0033] Specifically, as shown in Figure 6 The specific steps of obtaining the telemetry response accuracy index of the laser telemetry optical machine to be adjusted are as follows: obtaining the three-dimensional position coordinates of the detection of the laser telemetry optical machine to be adjusted (i.e. the three-dimensional position coordinates of the center point of the micro-displacement adjustment platform 703), and inputting the three-dimensional position coordinates of the detection of the laser telemetry optical machine to be adjusted and the electrical signal data into the pre-trained position optimization model for optimization analysis, obtaining the optimized three-dimensional position coordinates of the detection of the laser telemetry optical machine to be adjusted (indicating that the signal strength of the photodetector 701 is the largest and most stable when the micro-displacement adjustment platform 703 is at this position), and combining the three-dimensional position coordinates of the detection for comprehensive analysis (calculating the deviation between the three-dimensional position coordinates of the micro-displacement adjustment platform 703 and the optimized three-dimensional position coordinates based on the Euclidean distance formula, and mapping the result to 0-1 through the Sigmoid function), obtaining the detection position deviation index of the laser telemetry optical machine to be adjusted; based on the electrical signal data of the laser telemetry optical machine to be adjusted, analyzing the signal response index of the laser telemetry optical machine to be adjusted, and combining the position deviation index thereof, analyzing the telemetry response accuracy index of the laser telemetry optical machine to be adjusted.

[0034] The specific steps of analyzing the signal response index of the laser telemetry optical machine to be adjusted are as follows: performing time domain sampling processing on the electrical signal data of the laser telemetry optical machine to be adjusted, obtaining the signal amplitude of a plurality of time points of the laser telemetry optical machine to be adjusted, and performing comprehensive analysis on the signal amplitude of each time point, obtaining the signal strength index (i.e. the mean value of the signal amplitude of each time point), the signal stability index (i.e. the standard deviation of the signal amplitude of each time point), and the signal fluctuation index (i.e. performing statistical processing on the signal amplitude of each time point to obtain the maximum signal amplitude and the minimum signal amplitude, and performing difference processing, i.e. the absolute value of the difference between the maximum signal amplitude and the minimum signal amplitude), and performing standardization processing, and performing comprehensive analysis (i.e. weighted processing, and mapping the result to 0-1 through the Sigmoid function) on the signal strength index, the signal stability index, and the signal fluctuation index of the laser telemetry optical machine to be adjusted after standardization processing, obtaining the signal response index of the laser telemetry optical machine to be adjusted.

[0035] The position optimization model is specifically a convolutional neural network, including an input layer, a convolutional layer, a fully connected embedding layer, a fully connected fusion layer, and an output prediction layer. The specific steps for obtaining the optimized three-dimensional position coordinates of the detection of the laser remote sensing optical machine to be adjusted are as follows: in the input layer of the convolutional neural network, the three-dimensional position coordinates of the detection of the laser remote sensing optical machine to be adjusted and the electrical signal data are received and preprocessed; in the convolutional layer of the convolutional neural network, one-dimensional convolutional processing is performed on the preprocessed electrical signal data (photodetector 701) of the laser remote sensing optical machine to be adjusted (that is, a convolutional kernel with a step is used for feature extraction and dimension reduction to be performed synchronously, and a feature map compression mechanism is embedded in the convolutional layer to directly convert the multi-channel feature map output by convolution into a fixed-length signal feature vector), to obtain the detection signal feature vector of the laser remote sensing optical machine to be adjusted; in the fully connected embedding layer of the convolutional neural network, nonlinear mapping processing is performed on the preprocessed three-dimensional position coordinates of the detection of the laser remote sensing optical machine to be adjusted (the three-dimensional coordinates are received, normalized and standardized, and then input into a fully connected neural network including at least one layer with a nonlinear activation function, and after being weighted by weights, adjusted by a bias term, and converted by an activation function, a high-dimensional embedding feature vector is output to represent the spatial position), to obtain the detection space embedding vector of the laser remote sensing optical machine to be adjusted; in the fully connected fusion layer of the convolutional neural network, the detection feature vector and the detection space embedding vector of the laser remote sensing optical machine to be adjusted are fused (the signal feature vector and the space embedding vector are spliced in the feature dimension, that is, they are combined in the order of vector channels to form a unified fusion input vector, the fusion input vector contains the feature performance of the electrical signal received by the current detector in the time domain, such as intensity mode and fluctuation trend, and the position information of the current detector in the three-dimensional coordinate system, has cross-dimensional expression capability, and the fused feature vector is input into a fusion analysis module, which is composed of a group of nonlinear fully connected networks, usually including one or more neuron structures with activation functions, for realizing high-dimensional interaction learning and feature reconstruction between different source features, in the process, the fusion analysis layer not only models the potential nonlinear relationship between the signal and the position, but also compresses redundant or weakly correlated features, thereby outputting a signal-position joint semantic vector with a compact structure and high expression capability, which is marked as a comprehensive state feature vector), to obtain the detection comprehensive state feature vector of the laser remote sensing optical machine to be adjusted;In the output prediction layer of the convolutional neural network, the detection comprehensive state feature vector of the laser remote sensing optical machine to be adjusted is predicted (the output prediction layer is composed of fully connected neurons dedicated to regression analysis, used to learn the nonlinear mapping relationship between the comprehensive state feature and the optimal detection position, i.e., three output nodes are constructed for the weighted calculation path corresponding to the three direction spatial coordinate axes, each node performs linear weighted summation on each dimension feature in the input feature vector and adds a bias term to form a prediction output value, because the target is a continuous coordinate in the actual physical space, therefore the output layer does not use an activation function, and directly outputs a three-dimensional position prediction result in a linear manner, which is the position coordinate of the optimal signal receiving point under the current state), and the detection optimized three-dimensional position coordinate of the laser remote sensing optical machine to be adjusted is obtained.

[0036] The input layer is used to receive and pre-process the three-dimensional coordinates and electrical signal data of the photodetector.

[0037] The convolutional layer is used to extract the time domain features of the electrical signal and convert them into fixed-length feature vectors.

[0038] The fully connected embedding layer is used to map the three-dimensional position coordinates into a spatial embedding vector.

[0039] The fully connected fusion layer is used to fuse the signal features and the spatial embedding vector, and output a comprehensive state feature vector.

[0040] The output prediction layer is used to predict the optimized three-dimensional position coordinates based on the comprehensive feature vector.

[0041] The pre-training process of the convolutional neural network is as follows:

[0042] Sample data sets are obtained, including different spatial position coordinates, electrical signal sequences, and corresponding optimal three-dimensional position labels, and are divided into sample training sets and sample validation sets.

[0043] The convolutional neural network is initialized, including initializing the weight parameters and bias parameters of each neuron connection in the convolutional neural network structure, wherein the weight parameters are assigned values using a random initialization method based on a Gaussian distribution or a uniform distribution, and the bias parameters are usually initialized to zero or a small perturbation value, used to adjust the activation starting point of each neuron, enhance the response sensitivity of the nonlinear activation function, in addition, He initialization or Xavier initialization strategy can be selected according to the specific depth of the network structure and the type of activation function (such as ReLU, Tanh) to improve the convergence stability and efficiency of the model in the early stage of training.

[0044] Based on the sample training set, set the number of training cycles (such as 50 or 100), and in each training cycle, sequentially perform forward propagation (input the sample into the network, pass through each layer structure, and calculate the predicted three-dimensional coordinate output), loss calculation (compare the predicted coordinates with the real labeled coordinates, and calculate the error such as mean square error), back propagation (calculate the gradient of each layer parameter based on the error), and parameter update (use the optimizer such as Adam to update the weights and biases in the network).

[0045] After each training cycle, based on the sample validation set, perform evaluation analysis, input all samples in the validation set into the current trained convolutional neural network one by one, and obtain the corresponding predicted three-dimensional position coordinate output; compare the predicted output with the real labeled coordinates in the validation set, calculate the average error index (such as mean square error MSE, mean absolute error MAE, or Euclidean distance error) as the validation index after this round of training, at the same time, record the current validation error, and compare it with the historical optimal validation error, if the current error is better than the historical optimal value, update and save the model parameters as the current version; if the validation error does not decrease significantly for continuous multiple training cycles (i.e. meets the set "early stopping" judgment condition), or the validation error reaches the preset convergence threshold (such as error less than 0.001), the training process is terminated in advance, to avoid model overfitting to the training set, and ensure that the final model has good generalization ability.

[0046] The specific formula for calculating the remote sensing response accuracy index of the laser remote sensing optical machine to be adjusted is as follows: ; wherein, is the remote sensing response accuracy index of the laser remote sensing optical machine to be adjusted, is the signal response index of the laser remote sensing optical machine to be adjusted, is the response adjustment coefficient stored in the database, is the position deviation index of the laser remote sensing optical machine to be adjusted, is the position deviation adjustment coefficient stored in the database, is the interaction adjustment coefficient stored in the database.

[0047] It needs to be explained that, , , It can be obtained by the following steps: based on historical data, determine the initial influence weight of each variable (signal response index, position deviation index) on the remote sensing response accuracy index through statistical regression analysis, then adjust the value range of the coefficient using sensitivity analysis method to evaluate the stability and applicability of these parameters on the formula output, next, further fit the weight through model optimization (such as machine learning algorithm or multi-objective optimization) to ensure that the formula can accurately reflect the quality of the actual signal.

[0048] The specific implementation example of calculating the telemetry response accuracy index of the laser telemetry optical machine to be adjusted is as follows: the existing data includes the signal response index and the position deviation index of the laser telemetry optical machine to be adjusted after 5 times of random measurement analysis, and the specific implementation example is shown in Table 1.

[0049] Table 1: Index sequence data example of the laser telemetry optical machine to be adjusted

[0050] Signal response index Position deviation index First measurement 0.687 0.371 Second measurement 0.891 0.103 Third measurement 0.799 0.167 Fourth measurement 0.578 0.493 Fifth measurement 0.866 0.138

[0051] The response adjustment coefficient stored in the database is about 0.428.

[0052] The position deviation adjustment coefficient stored in the database is about 0.284.

[0053] The interaction adjustment coefficient stored in the database is about 1.462.

[0054] The data in Table 1 and the above adjustment coefficients are substituted into the specific formula for calculating the telemetry response accuracy index of the laser telemetry optical machine to be adjusted to obtain:

[0055] The telemetry response accuracy index of the laser telemetry optical machine to be adjusted after the first measurement analysis is about 0.667.

[0056] The telemetry response accuracy index of the laser telemetry optical machine to be adjusted after the second measurement analysis is about 1.172.

[0057] The telemetry response accuracy index of the laser telemetry optical machine to be adjusted after the third measurement analysis is about 0.819.

[0058] The telemetry response accuracy index of the laser telemetry optical machine to be adjusted after the fourth measurement analysis is about 0.599.

[0059] The telemetry response accuracy index of the laser telemetry optical machine to be adjusted after the fifth measurement analysis is about 1.102.

[0060] In the embodiment, by acquiring the electrical signal data of the photodetector 701 and the position data of the micro-displacement adjustment platform 703, and combining the real-time analysis of the position optimization model, the system can dynamically adjust the position of the detector to ensure maximum signal reception and stability, thereby significantly improving the measurement accuracy of the telemetry system in complex environments, especially in long-distance or high-interference conditions, effectively reducing measurement errors caused by position deviation or signal attenuation, and by optimizing the three-dimensional position coordinates, the detector is at the optimal signal receiving point, and the telemetry response accuracy is significantly improved. Secondly, the convolutional neural network is used for deep learning and feature extraction of the electrical signal data and the three-dimensional position coordinates, to automatically realize accurate matching of the signal and the position, and each training process can not only learn from historical data, but also optimize the network weight through multiple iterations, thereby realizing real-time adjustment, so that the system can adapt to complex working conditions, thereby improving the stability and reliability of the telemetry system, and then improving the overall work efficiency. Finally, the combination of signal response accuracy index and position deviation index provides a comprehensive telemetry evaluation index, so that the system can adjust the position of the detector according to the current measurement state, so that the system is not only suitable for static targets, but also can adapt to real-time changing measurement tasks, thereby making the telemetry optical machine better adapt to the scanning requirements of different targets, and ensuring the measurement accuracy.

[0061] Specifically, as Figure 7As shown, the component position data set includes the laser 101 (i.e. the emission port center point coordinates), the one-letter lens 201 (the center point coordinates), the aspheric collimating lens 203 (the center point coordinates), the receiving lens 3 (the center point coordinates), the target area (two ends) boundary point and the target center point three-dimensional position coordinates, and the specific steps of obtaining the structural optical matching index of the laser remote sensing light machine to be adjusted are as follows: based on the component position data set of the laser remote sensing light machine to be adjusted, the spot shape adaptation index and the light path alignment index of the laser remote sensing light machine to be adjusted are analyzed; the spot shape adaptation index and the light path alignment index of the laser remote sensing light machine to be adjusted are analyzed (and the weights corresponding to the spot shape adaptation index and the light path alignment index can be obtained based on the chaotic optimization dynamic weight algorithm, that is, the initial weight values corresponding to the spot shape adaptation index and the light path alignment index are allocated, and the sum of the two is 1, the initial weight can be set to be in a balanced state, such as both being 0.5, indicating that the initial contributions of the two parameters to the matching index are the same, the weight is updated iteratively by using the Logistic chaotic mapping formula, the chaotic control parameters are set to ensure that it is in a completely chaotic state, according to the weight of the current spot shape adaptation index, the weight of the next iteration step is calculated by the Logistic chaotic mapping formula, and the weight of the light path alignment index is updated synchronously, this process makes the weight randomly evolve in the interval of 0-1 through the ergodicity of chaotic mapping, covering all possible weight combinations, in each weight iteration, the structural optical matching index under the current weight is calculated, which comprehensively reflects the matching degree of the spot shape and the light path alignment, the higher the value, the better the system performance, within a preset number of iterations, such as 100 times, the value of the structural optical matching index of each iteration is recorded, and the weight combination that makes the index reach the maximum value is selected as the final fusion weight), and the structural optical matching index of the adjusted remote sensing light machine is obtained.

[0062] The specific steps of analyzing the spot shape adaptation index of the laser remote sensing light machine to be adjusted are as follows: the three-dimensional position coordinates of the laser 101 of the laser remote sensing light machine to be adjusted are comprehensively analyzed with the three-dimensional position coordinates of the one-letter lens 201 and the three-dimensional position coordinates of the aspheric collimating lens 203, to obtain a first distance value (i.e. the distance value between the laser 101 and the one-letter lens 201) and a second distance value (i.e. the distance value between the laser 101 and the aspheric collimating lens 203) of the laser remote sensing light machine to be adjusted, and statistical analysis (i.e. comparing and analyzing the first distance value and the second distance value to obtain the minimum distance value) is performed, to obtain the current use lens (if the first distance value is the minimum, the one-letter lens 201 is marked as the current use lens, if the second distance value is the minimum, the aspheric collimating lens 203 is marked as the current use lens) and the current distance value (if the current use lens is the one-letter lens 201, the first distance value is marked as the current distance value, if the current use lens is the aspheric collimating lens 203, the second distance value is marked as the current distance value) of the laser remote sensing light machine to be adjusted; and the aperture width of the laser 101 of the laser remote sensing light machine to be adjusted (which can be obtained from the equipment technical specification book stored in the database) and the laser divergence angle of the current use lens (which can be obtained from the calibrated divergence angle of the equipment stored in the optical database) are obtained, and are comprehensively analyzed in combination with the current distance value, to obtain the current standard spot width value of the laser remote sensing light machine to be adjusted, and the calculation formula is as follows: ; wherein, is the current standard spot width value of the laser remote sensing light machine to be adjusted, is the aperture width of the laser 101 of the laser remote sensing light machine to be adjusted, is the current distance value of the laser remote sensing light machine to be adjusted, is the laser divergence angle of the current use lens of the laser remote sensing light machine to be adjusted;

[0063] The target region (both ends) boundary point three-dimensional position coordinates of the laser remote sensing light machine to be adjusted are analyzed based on the Euclidean distance formula, to obtain the target width value of the laser remote sensing light machine to be adjusted, and the current standard spot width value is comprehensively analyzed in combination (i.e. the absolute value of the difference between the target width value and the current standard spot width value / the current standard spot width value, and then 1-the result), to obtain the spot shape adaptation index of the laser remote sensing light machine to be adjusted, and the calculation formula is as follows: ; wherein, is the spot shape adaptation index of the laser remote sensing light machine to be adjusted, is the target width value of the laser remote sensing light machine to be adjusted, is the current standard spot width value of the laser remote sensing light machine to be adjusted.

[0064] The specific steps of analyzing the light path alignment index of the laser remote sensing optical machine to be adjusted are as follows: obtaining the three-dimensional position coordinates of the current use lens of the laser remote sensing optical machine to be adjusted (three-dimensional position coordinates of the current use lens); reading the three-dimensional position coordinates of the laser 101, the receiving lens 3 and the target center point of the laser remote sensing optical machine to be adjusted, and combining the three-dimensional position coordinates of the current use lens to perform comprehensive analysis respectively, to obtain the emission direction vector of the laser remote sensing optical machine to be adjusted (i.e. three-dimensional vector subtraction processing based on the three-dimensional position coordinates of the laser 101 and the three-dimensional position coordinates of the current use lens), the receiving direction vector (i.e. three-dimensional vector subtraction processing based on the three-dimensional position coordinates of the receiving lens 3 and the three-dimensional position coordinates of the target center point), and similarity analysis, to obtain the light path alignment index of the laser remote sensing optical machine to be adjusted.

[0065] The specific formula for calculating the remote sensing credibility index of the laser remote sensing optical machine to be adjusted is as follows: ; wherein, is the remote sensing credibility index of the laser remote sensing optical machine to be adjusted, is the remote sensing response accuracy index of the laser remote sensing optical machine to be adjusted, is the signal quality adjustment coefficient stored in the database, is the structural optical matching index of the laser remote sensing optical machine to be adjusted, is the optical matching adjustment coefficient stored in the database, is the difference adjustment coefficient stored in the database, is the superposition adjustment coefficient stored in the database.

[0066] It needs to be explained that in the formula, This item is used to suppress the difference between the remote sensing response accuracy index and the structural optical matching index, to avoid the remote sensing credibility index being too high or too low.

[0067] , , , The initial weight value can be obtained by the following steps: using historical data, combining the remote sensing response accuracy index and the structural optical matching index, performing statistical regression analysis, quantifying the specific influence of each factor on the remote sensing credibility index, and then fitting out the initial weight value, and then using the sensitivity analysis method to adjust the value range of each coefficient, observing its influence on the remote sensing credibility evaluation result, to ensure the stability and rationality of the model.

[0068] In this embodiment, the system can dynamically adjust the shape and path alignment of the light spot to ensure optimal signal reception and maximum measurement accuracy through comprehensive analysis of the light spot shape adaptation index and the light path alignment index. The light spot shape adaptation index controls the switching between lenses to output different light spot shapes, effectively addressing different measurement needs. The light path alignment index ensures the accuracy of the light path alignment by precisely calculating the similarity of the light path to ensure the optimal docking of the laser beam and the receiving lens, thereby improving signal quality and reliability. In addition, the system can comprehensively evaluate signal quality and optical matching by combining the telemetry response accuracy index, the structural optical matching index, and the telemetry reliability index, dynamically adjusting the working state of each adjustment unit to improve the adaptability of the telemetry system in complex environments and effectively improve the stability and reliability of the system. Especially in long-distance telemetry or complex reflection scenarios, the system can adjust in real time to ensure optimal working state at all times. Finally, through automated data acquisition, analysis, and optical-mechanical adjustment, the system reduces human intervention, automatically adjusts to the best measurement state in complex environments, and reduces human error, thereby improving measurement accuracy and efficiency, making the system more suitable for remote operation and dynamic monitoring, especially in difficult-to-access scenarios, greatly improving the safety and efficiency of telemetry work.

[0069] Specifically, the specific steps of intelligently regulating and controlling the displacement adjustment unit 102, the detection light lens adjustment unit 2, and the micro-displacement adjustment platform 703 of the laser remote sensing light machine to be adjusted based on the remote sensing trust index are as follows: the remote sensing trust index of the laser remote sensing light machine to be adjusted is analyzed and judged with the preset remote sensing trust index threshold value; if the remote sensing trust index of the laser remote sensing light machine to be adjusted is higher than the preset remote sensing trust index threshold value, no regulation and control is performed; if the remote sensing trust index of the laser remote sensing light machine to be adjusted is lower than or equal to the preset remote sensing trust index threshold value, the displacement adjustment unit 102, the detection light lens adjustment unit 2, and the micro-displacement adjustment platform 703 of the laser remote sensing light machine to be adjusted are regulated and controlled, which is specifically that for the displacement adjustment unit 102, the displacement adjustment unit 102 is controlled by the signal processing unit 6, so that the laser 101 performs forward and backward step-by-step micro-movement (such as 0.1 mm as a unit), and the signal improvement amplitude is monitored; if the signal response index is obviously improved after adjustment, the current displacement value is locked as the focal length optimization parameter, if the signal improvement is not obvious, the adjustment is continued step by step until the optimal spot size is found, that is, in the axial micro-movement regulation process of the displacement adjustment unit 102, the system calculates the current signal response index in real time and compares it with the initial signal response index to obtain the improvement rate, and the system judges the regulation result according to the improvement rate: when the improvement rate is higher than or equal to 10%, it is considered that the signal improvement is significant, and the current displacement position is automatically locked; when the improvement rate is in [3%, 10%)], the next step-by-step micro-movement is continued; when it is lower than 3%, reverse adjustment or parameter rollback is performed; for the detection light lens adjustment unit 2, the rotary motor 204 is started to automatically switch the current lens state, that is, the spot shape adaptation index is analyzed and judged with the preset spot shape adaptation index threshold value, if the spot shape adaptation index is higher than the preset spot shape adaptation index threshold value, no switching is performed, and if it is lower, the rotary motor 204 is started to switch other shapes; for the micro-displacement adjustment platform 703, the detection optimization three-dimensional position coordinates of the laser remote sensing light machine to be adjusted are read, and the micro-displacement adjustment platform 703 is controlled by the signal processing unit 6 to move to the detection optimization three-dimensional position coordinates based on the detection optimization three-dimensional position coordinates, so as to adjust the position of the photodetector 701.

[0070] In the embodiment, the intelligent analysis of the telemetry credibility index dynamically determines when to adjust each unit in the system, so as to ensure timely fine-tuning in the case of poor signal quality, especially when the signal response index is not ideal, and the laser focal length is adjusted by step-by-step fine adjustment, so that the system can obtain the optimal spot size in real-time feedback, thereby improving the signal strength and stability, and the real-time monitoring of the signal improvement amplitude ensures that each adjustment can maximize the optimization of system performance, thereby avoiding unnecessary adjustment and improving overall accuracy. Secondly, the system reduces the intervention of manual operation through automatic spot adjustment and optical path optimization, and adjusts the micro-displacement adjustment platform, the detection light lens adjustment unit and the displacement adjustment unit through intelligent adjustment, so that the telemetry system can automatically adjust under different environmental conditions, not only improving the operation efficiency, but also ensuring the continuous stability under complex environment, thereby improving the adaptability of the system under dynamic working conditions. Finally, based on the analysis of the signal response index and the spot shape adaptation index in real-time feedback, the system can accurately adjust the spot shape and optical path alignment state to ensure that the system state after each adjustment is in the optimal range, so that the system can realize the most accurate signal reception in different optical path environments, effectively prevent the decline of telemetry accuracy, ensure long-term stable operation, and enhance the reliability of measurement.

[0071] Although preferred embodiments of the application have been described, those skilled in the art will be able to make additional changes and modifications to these embodiments once they have the basic inventive concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the application.

[0072] Obviously, those skilled in the art can make various modifications and variations to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalent technologies, the present application also intends to include these modifications and variations.

Claims

1. A laser remote sensing optical system with adjustable spot size, characterized in that, The system comprises: a laser emitting unit (1) for emitting a laser beam; a detection light lens adjusting unit (2) for receiving the laser beam emitted by the laser emitting unit (1) and performing shape control; a receiving lens (3) for receiving the light signal returned by the laser beam after reaching the target after the shape control of the detection light lens adjusting unit (2) and focusing; a detection unit (7) for converting the light signal focused by the receiving lens (3) into an electrical signal; a signal processing unit (6) for receiving the electrical signal output by the detection unit (7) and analyzing to generate a control instruction, the signal processing unit (6) comprises: a data acquisition module for acquiring state data of the laser remote sensing optical machine to be adjusted in real time, the state data comprising electrical signal data and component position data set; a data analysis module for performing feature analysis on the state data of the laser remote sensing optical machine to be adjusted respectively to obtain a remote sensing evaluation index set of the laser remote sensing optical machine to be adjusted, including a remote sensing response accuracy index, a structure optical matching index, and performing comprehensive analysis to obtain a remote sensing credibility index of the laser remote sensing optical machine to be adjusted; an optical machine adjusting module for intelligently controlling the laser emitting unit (1), the detection light lens adjusting unit (2) and the micro-displacement adjusting platform (703) of the laser remote sensing optical machine to be adjusted based on the remote sensing credibility index; the electrical signal data is specifically the signal amplitude at each time point, and the specific steps for obtaining the remote sensing response accuracy index of the laser remote sensing optical machine to be adjusted are as follows: acquire the detection three-dimensional position coordinates of the laser remote sensing optical machine to be adjusted, and input them and the electrical signal data into a pre-trained position optimization model for optimization analysis to obtain the detection optimized three-dimensional position coordinates of the laser remote sensing optical machine to be adjusted, and perform comprehensive analysis combined with the detection three-dimensional position coordinates to obtain the detection position deviation index of the laser remote sensing optical machine to be adjusted; based on the electrical signal data of the laser remote sensing optical machine to be adjusted, analyze the signal response index of the laser remote sensing optical machine to be adjusted, and combine it with the position deviation index to analyze the remote sensing response accuracy index of the laser remote sensing optical machine to be adjusted; the specific formula for calculating the remote sensing response accuracy index of the laser remote sensing optical machine to be adjusted is as follows: ； wherein, , , are, in sequence, a telemetry response accuracy index, a signal response index, a position deviation index of the laser telemetry optical machine to be adjusted, , , are the response adjustment coefficient, the position deviation adjustment coefficient, the interaction adjustment coefficient stored in the database.

2. The laser remote sensing optical machine system according to claim 1, wherein the detection unit (7) comprises a photodetector (701) for receiving the focused light signal of the receiving lens (3), and is stably carried through a detector mounting unit (702), and the photodetector (701) and the detector mounting unit (702) are positionally adjusted through a micro-displacement adjusting platform (703).

3. The laser range finder system according to claim 1, wherein The detection light lens adjusting unit (2) comprises a one-letter lens (201) for adjusting the shape of the laser beam emitted by the laser emitting unit (1) and an aspherical collimating lens (203), and the one-letter lens (201) and the aspherical collimating lens (203) are switched through a lens adjusting frame (202).

4. The laser range finder system according to claim 1, wherein The position optimization model is specifically a convolutional neural network, comprising an input layer, a convolutional layer, a fully connected embedding layer, a fully connected fusion layer and an output prediction layer, and the specific steps for obtaining the detection optimized three-dimensional position coordinates of the laser remote sensing optical machine to be adjusted are as follows: In the input layer of the convolutional neural network, the detection three-dimensional position coordinates and electrical signal data of the laser remote sensing optical machine to be adjusted are received and preprocessed; In the convolution layer of the convolutional neural network, one-dimensional convolution processing is performed on the preprocessed electrical signal data of the laser remote sensing optical machine to be adjusted, to obtain a detection signal feature vector of the laser remote sensing optical machine to be adjusted; In the fully connected embedding layer of the convolutional neural network, nonlinear mapping processing is performed on the preprocessed detection three-dimensional position coordinates of the laser remote sensing optical machine to be adjusted, to obtain a detection space embedding vector of the laser remote sensing optical machine to be adjusted; In the fully connected fusion layer of the convolutional neural network, the detection feature vector and the detection space embedding vector of the laser remote sensing optical machine to be adjusted are fused, to obtain a detection comprehensive state feature vector of the laser remote sensing optical machine to be adjusted; In the output prediction layer of the convolutional neural network, prediction processing is performed on the detection comprehensive state feature vector of the laser remote sensing optical machine to be adjusted, to obtain an optimized detection three-dimensional position coordinate of the laser remote sensing optical machine to be adjusted.

5. The laser range finder system according to claim 1, wherein The component position data set includes three-dimensional position coordinates of a laser (101), a linear lens (201), an aspherical collimating lens (203), a receiving lens (3), a target region boundary point, and a target center point, and the specific steps for obtaining the structural optical matching index of the laser remote sensing optical machine to be adjusted are as follows: Based on the component position data set of the laser remote sensing optical machine to be adjusted, the spot shape adaptation index and the optical path alignment index of the laser remote sensing optical machine to be adjusted are analyzed; The spot shape adaptation index and the optical path alignment index of the laser remote sensing optical machine to be adjusted are weighted and analyzed, to obtain the structural optical matching index of the adjusted remote sensing optical machine.

6. The laser range finder system of claim 1, wherein the laser range finder system further comprises a laser beam expander disposed between the laser source and the laser beam steering system. The specific formula for calculating the remote sensing credibility index of the laser remote sensing optical machine to be adjusted is as follows: ； Wherein, , , In turn, the remote sensing reliability index, the remote sensing response accuracy index, and the structured optical matching index of the laser remote sensing light machine to be adjusted, , , , The signal quality adjustment coefficient, the optical matching adjustment coefficient, the difference adjustment coefficient, and the superposition adjustment coefficient stored in the database.

7. The laser range finder system according to claim 1, wherein Based on the remote sensing credibility index, the specific steps for intelligently regulating and controlling the displacement adjustment unit (102), the detection optical lens adjustment unit (2), and the micro-displacement adjustment platform (703) of the laser remote sensing optical machine to be adjusted are as follows: The remote sensing credibility index of the laser remote sensing optical machine to be adjusted is compared with a preset remote sensing credibility index threshold value; If the remote sensing credibility index of the laser remote sensing optical machine to be adjusted is higher than the preset remote sensing credibility index threshold value, no regulation and control is performed; If the remote sensing credibility index of the laser remote sensing optical machine to be adjusted is lower than or equal to the preset remote sensing credibility index threshold value, the laser emission unit (1), the detection optical lens adjustment unit (2), and the micro-displacement adjustment platform (703) of the laser remote sensing optical machine to be adjusted are regulated and controlled.

Images