Three-dimensional modeling method and system for lunar lava duct

Abstract

The invention belongs to the field of lunar exploration and in-situ resource utilization, and particularly discloses a lunar lava pipe three-dimensional modeling method and system, and the method comprises the steps: carrying out the fusion of multi-modal data of a lunar lava pipe, and obtaining a basic three-dimensional model; the multi-modal data comprises point cloud data collected by a laser radar, point cloud data collected by a surrounding device penetrating radar, a three-dimensional image collected by a stereoscopic vision camera and a thermal image collected by an infrared thermal imager; respectively inputting a complete area in the basic three-dimensional model into a generation model based on the CGAN and the VAE, and correspondingly obtaining two three-dimensional completion models; and carrying out weighted fusion on the two three-dimensional complementation models to obtain a complete three-dimensional model of the lunar lava pipeline. According to the method, the integrity and precision of the three-dimensional model of the lunar lava pipeline can be improved, and technical support is provided for deep detection of the lunar lava pipeline.

Description

Technical Field

[0001] The present invention belongs to the field of lunar exploration and in-situ resource utilization, and more specifically, relates to a three-dimensional modeling method and system for lunar lava tubes. Background Art

[0002] In recent years, as the lunar exploration mission has been carried out, lunar lava tubes have gradually attracted widespread attention from the scientific community. Lunar lava tubes are special underground caves formed by ancient volcanic activity on the moon. When the lava cools and solidifies, due to the speed difference, a solid insulation layer is formed on the outside, while the lava inside continues to flow, eventually forming a hollow tube structure. Due to the low gravity environment of the moon, the scale of lunar lava tubes is much larger than that of lava tubes on Earth, with a diameter of hundreds or even thousands of meters, making it an ideal site for a lunar base. These lava tubes can not only serve as a natural barrier to effectively block cosmic radiation, reduce the impact of lunar micrometeorites and lunar dust on equipment and personnel, but also may store rich geological information, providing valuable data for studying the internal structure of the moon.

[0003] However, the detection of lunar lava tubes faces multiple technical challenges. First, due to the unknown underground environment of the lava tubes, the detection equipment cannot rely on prior maps for navigation, and the pathfinding and modeling rates are greatly reduced; second, the harsh environment on the lunar surface, such as lunar dust and radiation, will interfere with wireless communication signals, making it difficult to maintain real-time data transmission and command interaction. In addition, the lack of satellite navigation signals such as GPS in the lunar lava tubes increases the difficulty of equipment navigation and positioning. Finally, the lighting in the lunar lava tubes is extremely limited, making it difficult for visual sensors to capture clear images, greatly reducing perception and positioning accuracy, and affecting the equipment's autonomous detection capabilities.

[0004] To address these challenges, multimodal data fusion technology has shown great potential in the 3D modeling of lunar lava tubes. By combining multiple sensors such as lidar, orbiter penetrating radar, stereo vision camera and infrared thermal imager, it is possible to capture complex terrain details more comprehensively and improve the accuracy of 3D modeling. However, in practical applications, how to effectively integrate and synchronously process data from different sensors and infer unscanned areas are still technical difficulties that need to be solved. Summary of the invention

[0005] In view of the above defects or improvement needs of the prior art, the present invention provides a method and system for three-dimensional modeling of lunar lava tubes, the purpose of which is to improve the integrity and accuracy of three-dimensional modeling of lunar lava tubes.

[0006] To achieve the above object, according to one aspect of the present invention, a method for three-dimensional modeling of lunar lava tubes is proposed, comprising the following steps:

[0007] The multimodal data of the lunar lava tubes are fused to obtain a basic three-dimensional model; the multimodal data includes point cloud data collected by a laser radar, point cloud data collected by an orbiter penetrating radar, three-dimensional images collected by a stereo vision camera, and thermal images collected by an infrared thermal imager;

[0008] The complete areas in the basic 3D model are respectively input into the generative models based on CGAN and VAE to obtain two corresponding 3D completion models; the two 3D completion models are weightedly fused to obtain a complete 3D model of the lunar lava tube.

[0009] As a further preferred method, the method for obtaining multimodal data is specifically as follows:

[0010] In the lunar lava tubes, the lidar uses short-wave infrared laser technology to collect point cloud data; the orbiter penetrating radar collects point cloud data through 30Hz to 100Hz VHF signals; the stereo vision camera uses digital light enhancement technology to generate three-dimensional images; and an infrared thermal imager with a sensitivity of 0.05℃ to 0.1℃ and a band of 3μm to 14μm is used to collect thermal images.

[0011] As a further preference, when acquiring multimodal data, auxiliary positioning is performed through an inertial navigation system.

[0012] As a further preferred method, the multimodal data of the lunar lava tubes are fused to obtain a basic three-dimensional model, including the following steps:

[0013] The point cloud data collected by the laser radar, the point cloud data collected by the surround penetrating radar, and the images collected by the stereo vision camera are reconstructed in three dimensions through the SLAM technology, and three corresponding three-dimensional reconstruction models are obtained;

[0014] Feature points are extracted from the three 3D reconstruction models and thermal images respectively, and then based on the feature points, the model is matched by ICP algorithm to obtain the basic 3D model.

[0015] As a further preferred method, the model matching is performed by using an ICP algorithm, specifically:

[0016] In each iteration of the ICP algorithm, when the error function value is less than the error threshold, the iteration is stopped to obtain the optimal transformation matrix and achieve model matching; the error function adopts the mean square error MSE;

[0017] The error threshold gradually decreases as the iteration proceeds:

[0018]

[0019] Among them, T nis the error threshold at the nth iteration, T0 is the initial error threshold, error(0) is the initial error, error(i) is the error at the ith iteration, and β is the parameter for adjusting the decay rate.

[0020] As a further preferred method, the method for constructing a generative model based on VAE includes:

[0021] Perform data enhancement on the basic 3D model to obtain a preliminary 3D model; extract several complete parts from the preliminary 3D model to construct a training set;

[0022] VAE includes an encoder and a decoder. VAE is trained through a training set: the encoder takes the complete part as input to obtain latent space variables; the decoder decodes the latent space variables to generate a three-dimensional completion model; the loss function L used during training VAE as follows:

[0023] L VAE =D KL (q(z|x)||p(z))-E q(z|x) [p(x|z)]

[0024] Among them, D KL represents KL divergence, q(z|x) represents the posterior distribution of the latent variable z generated by the encoder according to the input data x, p(z) represents the prior distribution of the latent variable z, and || represents the distribution difference operator in KL divergence; E q(z|x) represents the expectation of the latent variable z, and p(x|z) represents the three-dimensional completion model generated by the decoder;

[0025] The trained VAE is the VAE-based generative model.

[0026] As a further preferred method, the method for constructing a generative model based on CGAN includes:

[0027] CGAN consists of a generator and a discriminator. CGAN is trained through a training set: the generator takes the complete part as input to generate a 3D completion model; the discriminator determines the authenticity of the complete part and the 3D completion model generated by the generator; the loss function L used in training CGAN as follows:

[0028] L CGAN =E[logD(x)]+E[log(1-D(G(z))]

[0029] Where E represents the expectation of all samples, G(z) represents the three-dimensional completion model generated by the generator, D is the discriminator, which determines the probability of the data being true, D(x) represents the judgment result of the input data x, and D(G(z)) represents the judgment result of the three-dimensional completion model G(z);

[0030] The trained generator is the CGAN-based generation model.

[0031] As a further preferred embodiment, data enhancement is performed on the basic three-dimensional model, including: data enhancement is performed on the basic three-dimensional model by adding noise, local cropping and local magnification.

[0032] As a further preferred method, the two three-dimensional completion models are weightedly fused, including: using a multilinear interpolation method to weightedly fuse the two three-dimensional completion models; and then correcting the geometric consistency of the fused models through a global optimization algorithm to obtain a complete three-dimensional model of the lunar lava tube.

[0033] According to another aspect of the present invention, a lunar lava tube three-dimensional modeling system is provided, comprising a processor, wherein the processor is used to execute the above-mentioned lunar lava tube three-dimensional modeling method.

[0034] In general, the above technical solution conceived by the present invention has the following technical advantages compared with the prior art:

[0035] 1. The present invention combines the 3D modeling method of multimodal data fusion with generative model inference and completion to infer and complete the structure of the unscanned area of ​​the lunar lava tube; thereby improving the integrity and accuracy of the constructed 3D model of the lunar lava tube, and providing important technical reserves and scientific basis for future lunar lava tube exploration.

[0036] 2. The present invention optimizes the acquisition of multimodal sensor data to address the challenges of detecting no light, complex terrain, and deep structures in lunar lava tubes. The laser radar uses short-wave infrared laser technology to reduce the interference of lunar dust reflection and light scattering, and enhance the detection capability of the pipeline surface and deep structures. The surround penetrating radar uses the low VHF band and has a strong penetration detection capability. It can obtain the overall direction and hidden deep structure of the lunar lava tube, and provide accurate pipeline model positioning and direction information. The stereo vision camera uses digital illumination enhancement technology, which is suitable for the weak light environment of the lunar lava tube, and can generate high-precision stereo images in a light-limited environment. The infrared thermal imager can capture local small temperature differences in a large temperature difference environment through high sensitivity adjustment, and identify cracks, rock changes and hidden structures inside the pipeline. The collaborative fusion of multiple sensors ensures that the system can achieve high-precision detection and three-dimensional modeling in the complex environment of the moon.

[0037] 3. The present invention integrates an inertial navigation system (INS), which can achieve high-precision positioning even in a signal-free environment, ensuring the navigation and modeling accuracy of the equipment in the lunar lava tubes.

[0038] 4. When dealing with complex terrain or occlusion, a single sensor has limited perception capabilities and is unable to provide comprehensive data support. The present invention achieves accurate modeling of the internal structure of complex lunar lava tubes through multi-sensor data fusion. The laser radar and the surround penetrating radar provide three-dimensional point clouds and information on the overall direction of the pipeline, the stereo vision camera generates a three-dimensional depth map, and the infrared thermal imager supplements the temperature change data; then the SLAM technology is used to generate a three-dimensional point cloud model of the scanned area, and based on the ICP algorithm, the multimodal data of the lunar lava tube is aligned in time and space to improve the robustness of data fusion. This fusion technology overcomes the limitations of a single sensor in complex terrain, enhances data comprehensiveness, improves model accuracy and stability, and meets the needs of three-dimensional modeling of lunar lava tubes.

[0039] 5. In lunar lava tubes, some areas are difficult to fully scan due to narrow structures and occlusions, resulting in missing models. The present invention uses a pre-trained generative large model to infer and complete the unscanned areas of lunar lava tubes. Specifically, a generative large model combining variational autoencoders (VAE) and conditional generative adversarial networks (CGAN) is used to infer and complete the three-dimensional structure of the unscanned area. VAE encodes the three-dimensional point cloud of the scanned area, extracts complex geometric features, and reconstructs the spatial information of the missing area through decoding; CGAN infers the three-dimensional structure of the unscanned area from the scanned data through adversarial training of the generator and the discriminator, and improves the performance of the generated three-dimensional model in terms of details and overall consistency through multiple iterative training and dynamic parameter challenges. This model completion method further improves the modeling accuracy of the unscanned area through multi-source data fusion, ensures the integrity of the three-dimensional model in a complex environment, and meets the modeling requirements of lunar lava tubes. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 This is a schematic diagram of the process of the three-dimensional modeling method of a lunar lava tube according to an embodiment of the present invention;

[0041] Figure 2 A schematic diagram of the basic three-dimensional model construction process of an embodiment of the present invention;

[0042] Figure 3 The figure is a schematic diagram of the three-dimensional model completion optimization process according to an embodiment of the present invention. DETAILED DESCRIPTION

[0043] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0044] The embodiment of the present invention provides a method for three-dimensional modeling of lunar lava tubes, such as Figure 1 As shown, the following steps are included:

[0045] S1. Multimodal data acquisition and preliminary processing: Use lidar, orbiter penetrating radar, stereo vision camera and infrared thermal imager to collect multimodal data of lunar lava tubes.

[0046] Furthermore, the laser radar generates high-precision three-dimensional point cloud data through short-wave infrared lasers, reduces the interference of lunar dust and reflected light, and ensures accurate detection of complex terrain such as bulges, collapses and intersections of lunar lava tubes. The orbiter penetrating radar penetrates the lunar surface through low VHF signals (preferably 30Hz to 100Hz), captures the direction, structure and deep hidden areas of the lunar lava tubes, and obtains rough three-dimensional morphological information of the lunar lava tubes. The stereo vision camera uses digital illumination enhancement technology to enhance the visual effect through active fill light and dark image preprocessing, ensuring that the geometric texture details of the lava tube surface are captured under dim or no light conditions, and high-resolution images are obtained. A high-sensitivity infrared thermal imager (preferably with a sensitivity of 0.05℃ to 0.1℃ and a band of 3μm to 14μm) is used to capture the temperature distribution in the tube, which can capture subtle temperature changes in the lunar lava tubes under extreme temperature differences, detect cracks, rock layer changes and other discontinuous structures in the lunar lava tubes, and identify the thermal radiation characteristics of different geological layers. After synchronous data collection by multiple sensors, the data is denoised and invalid data is eliminated to ensure data quality.

[0047] Specifically, the collection operation is carried out in the following steps: The orbiter penetrating radar is located on a satellite or orbiter platform, and uses low VHF signals to penetrate the lunar surface to detect the direction and deep structure of the lunar lava tube. At the same time, the detection equipment is mounted on a mobile platform and slowly moves along the lava tube, triggering multimodal sensors to collect synchronously at fixed intervals or at key geological structure locations. The lidar collects point cloud data in real time, and removes interference from lunar dust and reflected light through a built-in filtering system to ensure the clarity of the detection data. The binocular stereo images generated by the stereo vision camera can clearly reflect the details and geometric changes on the surface of the lava tube. The infrared thermal imager captures the temperature difference distribution inside the lava tube, identifies areas with large temperature differences and local hotspots, and helps locate areas with large terrain changes in the tube.

[0048] The collected multimodal data is transmitted wirelessly to the platform computing unit, and after format conversion and noise filtering, it is aligned in time and space according to the collected location information, timestamp, and trajectory of the mobile platform. During the alignment process, the platform computing unit adjusts the data of each sensor in real time to ensure that the multimodal data is accurately aligned in the same time axis and spatial coordinate system. At the same time, the data processing unit monitors the quality of each sensor data in real time, and triggers the resampling mechanism when detecting missing data or large errors to ensure data integrity.

[0049] Furthermore, the acquisition equipment has been optimized to adapt to the complex environment of the lunar lava tubes, and can work stably under conditions without communication signals, positioning navigation, and light. The data of each sensor is assisted in positioning by the inertial navigation system (INS), so that the equipment can achieve high-precision positioning and modeling operations through INS in an environment without navigation signals. The equipment has local data storage and automatic synchronization functions to ensure that the equipment can continuously collect and store multimodal data in the absence of communication signals; after the communication signal is restored, the system will automatically upload the stored data to achieve seamless data synchronization and ensure data integrity and continuity. In addition, the internal processing unit of the equipment has been optimized to ensure independent data processing and fusion without external communication, ensuring the smooth progress of the exploration mission.

[0050] S2. Multimodal data alignment and fusion: Based on the collected multimodal data, SLAM technology is used for three-dimensional reconstruction, and then the ICP algorithm is used to match and fuse the multimodal three-dimensional models to generate the basic three-dimensional model of the lava tube.

[0051] like Figure 2 As shown, the specific steps include:

[0052] S21. Use SLAM technology to perform three-dimensional reconstruction on the point cloud data collected by the laser radar, the point cloud data collected by the surround penetrating radar, and the images collected by the stereo vision camera, and obtain three corresponding three-dimensional reconstruction models.

[0053] Specifically, the laser radar data is processed by LIDAR-SLAM to generate a high-precision three-dimensional point cloud model; the orbiter penetrating radar data is processed by GPR-SLAM to obtain the overall direction and deep structure of the lava tube; the three-dimensional image of the stereo vision camera is processed by Stereo Visual SLAM to generate a three-dimensional surface structure map, thereby improving the texture information of the three-dimensional modeling of the lunar lava tube.

[0054] S22, extracting feature points from the three 3D reconstruction models and the thermal image respectively, and then performing model matching based on the feature points through the ICP algorithm to obtain a basic 3D model. Specifically including:

[0055] First, feature points are extracted from each model. Feature point extraction is based on normal vector and curvature analysis to select key points representing the geometric structure of the model.

[0056] Then the improved ICP algorithm is used to register the feature point set, and the registration error is calculated in the iterative process. The error calculation optimization index is the mean square error (MSE), which is expressed as follows:

[0057]

[0058] Among them, p i is the feature point of the source point cloud, q i is the target point cloud feature point, N is the number of feature points; R is the rotation matrix, and t is the translation vector. This error function describes the matching error between point clouds, and the optimization goal is to minimize the error by adjusting the rotation and translation matrices.

[0059] The above indicators are used to evaluate whether the model meets the set threshold standard. When the MSE drops to the error threshold, the transformation matrix at this time is output to complete the model alignment.

[0060] Each model is matched according to the above method. For example, first, the point cloud data collected by the lidar is matched with the point cloud data collected by the penetrating radar of the orbiter; then, the 3D model obtained by the above matching is aligned with the 3D image collected by the stereo vision camera; then, the aligned 3D model is matched with the thermal image data collected by the infrared thermal imager, and finally a basic 3D model of the lunar lava tube is formed. The model includes the terrain undulations, inner wall contours and height differences of the lunar lava tube.

[0061] Furthermore, the standard ICP algorithm uses a fixed threshold for point cloud matching, while the improved ICP algorithm of the present invention introduces an adaptive threshold adjustment mechanism based on error accumulation. Specifically, a larger error threshold is set at the beginning of the registration to achieve a rough match and quickly reduce the multimodal data error; as the iteration proceeds, the error threshold is gradually reduced according to the error change of each iteration, thereby improving the registration accuracy. The error threshold update formula is as follows:

[0062]

[0063] Among them, T n is the error threshold at the nth iteration, T0 is the initial error threshold, error(0) is the initial error, error(i) is the error at the ith iteration, and β is the parameter for adjusting the decay rate.

[0064] S3. Training and inference completion of generative large models: Due to occlusion and other factors in the data collection process, the basic 3D model is missing. Based on the generative model, the basic 3D model is optimized and completed to obtain a complete 3D model of the lunar lava pipe.

[0065] like Figure 3 As shown, the specific steps include:

[0066] S31. Pre-trained generative model.

[0067] First, a training set is constructed: data enhancement is performed on the basic 3D model, specifically using data enhancement techniques such as adding noise, local cropping, and local amplification to increase the diversity of the training set and improve the generalization ability of the model in inferring unknown areas, thereby obtaining a preliminary 3D model; then, a complete 3D structure is extracted from the preliminary 3D model to construct a training set. In some embodiments, samples of lava tubes on Earth can also be used to create a training set for model training.

[0068] Then, VAE and CGAN are trained respectively using the training set to obtain generation models based on VAE and CGAN.

[0069] Furthermore, the VAE model includes an encoder and a decoder structure. The encoder takes the complete part from the basic 3D model as input, extracts latent space variables, and focuses on the reconstruction of the missing area; the decoder decodes according to the latent space variables, infers the geometric shape of the missing area, and generates a completed 3D model. The VAE loss function during training is as follows:

[0070] L VAE =D KL (q(z|x)||p(z))-E q(z|x) [p(x|z)]

[0071] Among them, D KL represents KL divergence, q(z|x) represents the posterior distribution of the latent variable z generated by the encoder, p(z) represents the prior distribution of the latent variable z, and || represents the distribution difference operator in KL divergence; E q(z|x) represents the expectation of the latent variable z, p(x|z) represents the 3D completion model generated by the decoder; z represents the latent variable, x represents the input data or 3D model;

[0072] Furthermore, the CGAN model includes a generator and a discriminator structure. The generator takes the complete part of the basic model as input to generate a 3D completion model; the discriminator distinguishes the complete part from the 3D completion model, evaluates the authenticity of the generated result, and optimizes the generator performance through the loss function. The CGAN loss function during training is:

[0073] L CGAN=E[logD(x)]+E[log(1-D(G(z)))]

[0074] Among them, L CGAN is the CGAN loss function, E represents the expectation of all samples, G(z) represents the 3D completion model generated by the generator, D is the discriminator, which determines the probability of the data being real, D(x) represents the judgment result of the input data x, and D(G(z)) represents the judgment result of the 3D completion model G(z)

[0075] After training, the VAE and CGAN generators can complete the three-dimensional model respectively. The VAE generates the geometric shape of the missing area through latent space variables, and the CGAN generator generates complete three-dimensional data.

[0076] S32. The complete areas in the basic three-dimensional model are respectively input into the generation models based on CGAN and VAE, and two three-dimensional completion models are obtained accordingly. The occluded or unscanned areas are completed, and the two three-dimensional completion models are weightedly fused to obtain a complete three-dimensional model of the lunar lava tube.

[0077] Furthermore, the two 3D completion models are weightedly fused, including: using multilinear interpolation to weight the overlapping areas of the two models; and then correcting the geometric consistency of the fused models through a global optimization algorithm to obtain a complete 3D model. The complete 3D model has continuity and consistency, covering known areas and inferring unknown areas.

[0078] Specifically, weighted fusion assigns different weights according to the quality and confidence of the data to ensure that high-quality data has a greater impact on the final model; the weighted process uses multiple linear interpolation to ensure smooth transitions in the boundary area. After splicing, the geometric consistency is corrected through a global optimization algorithm to correct errors and discontinuities at the splicing point; the global error function is optimized by minimizing the following formula:

[0079]

[0080] Among them, p i With q i The corresponding point of the boundary area, T(q i ) is the geometric transformation of the boundary area.

[0081] It will be easily understood by those skilled in the art that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A three-dimensional modeling method for lunar lava tubes, characterized in that: The steps include: The multimodal data of the lunar lava tubes are fused to obtain a basic three-dimensional model; the multimodal data includes point cloud data collected by a laser radar, point cloud data collected by an orbiter penetrating radar, three-dimensional images collected by a stereo vision camera, and thermal images collected by an infrared thermal imager; The complete areas in the basic 3D model are respectively input into the generative models based on CGAN and VAE to obtain two corresponding 3D completion models; the two 3D completion models are weightedly fused to obtain a complete 3D model of the lunar lava tube.

2. The three-dimensional modeling method of lunar lava tube according to claim 1, characterized in that: The specific method for obtaining multimodal data is: In the lunar lava tubes, the lidar uses short-wave infrared laser technology to collect point cloud data; the orbiter penetrating radar collects point cloud data through 30Hz to 100Hz VHF signals; the stereo vision camera uses digital light enhancement technology to generate three-dimensional images; and an infrared thermal imager with a sensitivity of 0.05℃ to 0.1℃ and a band of 3μm to 14μm is used to collect thermal images.

3. The three-dimensional modeling method of lunar lava tube according to claim 1, characterized in that: When acquiring multimodal data, auxiliary positioning is performed through the inertial navigation system.

4. The method for three-dimensional modeling of lunar lava tubes according to claim 1, characterized in that: The multimodal data of lunar lava tubes are fused to obtain a basic three-dimensional model, including the following steps: The point cloud data collected by the laser radar, the point cloud data collected by the surround penetrating radar, and the images collected by the stereo vision camera are reconstructed in three dimensions through the SLAM technology, and three corresponding three-dimensional reconstruction models are obtained; Feature points are extracted from the three 3D reconstruction models and thermal images respectively, and then based on the feature points, the model is matched by ICP algorithm to obtain the basic 3D model.

5. The method for three-dimensional modeling of lunar lava tubes according to claim 4, characterized in that: Model matching is performed through the ICP algorithm, specifically: In each iteration of the ICP algorithm, when the error function value is less than the error threshold, the iteration is stopped to obtain the optimal transformation matrix and achieve model matching; the error function adopts the mean square error MSE; The error threshold gradually decreases as the iteration proceeds: Among them, T n is the error threshold at the nth iteration, T0 is the initial error threshold, error(0) is the initial error, error(i) is the error at the ith iteration, and β is the parameter for adjusting the decay rate.

6. The method for three-dimensional modeling of lunar lava tubes according to claim 1, characterized in that: The construction methods of VAE-based generative models include: Perform data enhancement on the basic 3D model to obtain a preliminary 3D model; extract several complete parts from the preliminary 3D model to construct a training set; VAE includes an encoder and a decoder. VAE is trained through a training set: the encoder takes the complete part as input to obtain latent space variables; the decoder decodes the latent space variables to generate a three-dimensional completion model; the loss function L used during training VAE as follows: L VAE =D KL (q(z|x)||p(z))-E q(z|x) [p(x|z)] Among them, D KL represents KL divergence, q(z|x) represents the posterior distribution of the latent variable z generated by the encoder according to the input data x, p(z) represents the prior distribution of the latent variable z, and || represents the distribution difference operator in KL divergence; E q(z|x) represents the expectation of the latent variable z, and p(x|z) represents the three-dimensional completion model generated by the decoder; The trained VAE is the VAE-based generative model.

7. The method for three-dimensional modeling of lunar lava tubes according to claim 6, characterized in that: The construction method of the CGAN-based generative model includes: CGAN consists of a generator and a discriminator. CGAN is trained through a training set: the generator takes the complete part as input to generate a 3D completion model; the discriminator determines the authenticity of the complete part and the 3D completion model generated by the generator; the loss function L used in training CGAN as follows: L CGAN =E[logD(x)]+E[log(1-D(G(z))] Where E represents the expectation of all samples, G(z) represents the three-dimensional completion model generated by the generator, D is the discriminator, which determines the probability of the data being true, D(x) represents the judgment result of the input data x, and D(G(z)) represents the judgment result of the three-dimensional completion model G(z); The trained generator is the CGAN-based generation model.

8. The method for three-dimensional modeling of lunar lava tubes according to claim 7, characterized in that: The basic 3D model is enhanced by data enhancement, including: enhancing the basic 3D model by adding noise, local cropping and local magnification.

9. The method for three-dimensional modeling of a lunar lava tube according to any one of claims 1 to 8, characterized in that: The two three-dimensional completion models are weightedly fused, including: using a multilinear interpolation method to weightedly fuse the two three-dimensional completion models; then correcting the geometric consistency of the fused model through a global optimization algorithm to obtain a complete three-dimensional model of the lunar lava tube.

10. A lunar lava tube 3D modeling system, characterized in that: It includes a processor, which is used to execute the lunar lava tube three-dimensional modeling method as described in any one of claims 1-8.

Images