Ros2-based face exposure rapid prototyping method, device, system and medium

Abstract

The ROS2-based surface exposure rapid prototyping method, device, system and medium relate to the field of additive manufacturing technology. The method comprises the following steps: zero initialization of each hardware node, the prototyping platform is flush with the powder laying plane, and the powder is supplemented and the scraper is reset. After the three-dimensional model of the protyped part is imported into the slicing software based on ROS2, the support structure is generated after posture optimization, the slicing layer is obtained according to the preset layer thickness, and the rasterization processing is performed to generate a black and white mask image. The subsequent steps are repeated for each slicing layer until all the slicing layers are completed: the prototyping bin servo motor is lowered by the preset layer thickness and the powder is provided by the powder supply bin, and then the powder is spread and scraped. The black and white mask image corresponding to the current slicing layer is sent to the DLP for projection. The SLM laser node performs selective laser melting according to the projected area. The molten pool image is captured in real time and analyzed, and the analysis result is input into the ROS2 decision node to dynamically adjust the process parameters, so that closed-loop control is realized.

Description

Technical Field

[0001] This invention relates to the field of additive manufacturing technology, and more specifically, to a method, apparatus, system, and medium for rapid prototyping based on ROS2 surface exposure. Background Technology

[0002] As product designs become increasingly complex, traditional manufacturing methods struggle to meet the demands of rapid prototyping and small-batch production. Therefore, metal laser additive manufacturing technology has garnered significant attention due to its advantage of not being limited by structural complexity. This technology enables the direct, layer-by-layer construction of parts based on digital models, thereby shortening development cycles and reducing costs. However, existing methods still face numerous challenges in terms of efficiency, accuracy, and stability.

[0003] Existing metal laser additive manufacturing technologies mainly include powder-spread selective laser melting (SLM) and powder-feed direct laser metal forming. SLM technology is based on a point light source scanning powder layers one by one to complete part construction. This method melts metal powder point by point with a laser beam, stacking layers to form a three-dimensional solid. It has been developed for over thirty years and has been applied in practice. While the point light source scanning method can handle complex structures, its core process relies on slow point-by-point movement, resulting in inherent limitations in the overall forming process.

[0004] However, existing technologies have significant drawbacks: point light source scanning efficiency is low, making it difficult to handle the production of large-sized parts. Forming accuracy and consistency are insufficient, and material deformation and residual stress are easily caused by heat accumulation. Quality monitoring and control are lagging, with defects such as porosity or cracks often only being discovered after forming is complete. Environmental adaptability is poor, with dust and heat interference affecting stability. Process fluctuations lead to inconsistent forming quality, increasing scrap rates and post-processing costs. These problems restrict the further industrial application of metal additive manufacturing. Summary of the Invention

[0005] The present invention provides a method, apparatus, system, and medium for rapid prototyping based on ROS2 surface exposure, in order to improve at least one of the above-mentioned technical problems.

[0006] In a first aspect, the present invention provides a surface exposure rapid prototyping method based on ROS2, which includes steps S1 to S4.

[0007] S1. Based on ROS2, each hardware node performs zero initialization, controls the molding platform to keep it level with the powder spreading plane, and completes powder replenishment and scraper reset.

[0008] S2. Import the 3D model of the part to be formed into ROS2-based slicing software, perform posture optimization and generate support structure, and slice according to the preset layer thickness to obtain the 2D contour of each slice layer.

[0009] S3. Rasterize the two-dimensional contours of each slice layer to generate the corresponding black and white mask images.

[0010] S4. Repeat steps S41 to S45 for each slice layer until all slice layers are completed:

[0011] S41. Control the servo motor of the forming chamber to descend according to the preset layer thickness, and supply powder from the powder supply chamber.

[0012] S42. Spread and level the powder on the forming platform to ensure that the powder layer is evenly distributed.

[0013] S43. Send the black and white mask image corresponding to the current slice layer to the DLP and drive the DLP auxiliary node to perform projection.

[0014] S44. Drive the SLM laser node to perform selective laser melting according to the projected area.

[0015] S45. Start the AI ​​visual monitoring node to capture and analyze the molten pool image in real time, and input the analysis results into the ROS2 decision node to dynamically adjust the process parameters and achieve closed-loop control.

[0016] As a further aspect of the present invention, S1 includes S11 to S13.

[0017] S11 and ROS2 drive the molding chamber servo motor to move to the zero position. If the molding platform and the powder spreading plane are not kept horizontally aligned, control the molding platform to lift up or lower down until it is kept horizontally aligned.

[0018] S12 and ROS2 drive the powder supply servo motor to control the powder supply hopper to descend to the powder filling position to complete powder replenishment, and then rise to the initial powder supply height.

[0019] S13 and ROS2 drive the scraper stepper motor control node, driving the scraper to return to the safe position.

[0020] As a further aspect of the present invention, S2 includes S21 to S24.

[0021] S21. Import the STL 3D model of the part to be formed.

[0022] S22. Automatically perform posture optimization on the STL 3D model.

[0023] S23. Generate a support structure for molding based on the optimized model.

[0024] S24. The STL three-dimensional model is sliced ​​based on the preset layer thickness parameters to obtain multiple slice layers, and the two-dimensional contour of each slice layer is obtained.

[0025] As a further aspect of the present invention, S3 includes S31 to S34.

[0026] S31. Rasterize the two-dimensional contours of each slice layer to generate corresponding black-and-white mask images, including:

[0027] S32. Map the two-dimensional contours of each slice layer to pixel coordinates to generate the initial black and white mask image.

[0028] S33. Determine whether the pre-printed pattern of a single slice layer exceeds the projection range of the device's maximum laser filling spot.

[0029] S34. If the projection range is exceeded, the initial black and white mask image is divided into multiple sub-regions in an M×N checkerboard pattern, and multiple corresponding sub-mask images are generated as images to be projected in sequence.

[0030] As a further embodiment of the present invention, S43 includes S431 to S432.

[0031] S431. If the pre-printed pattern of a single slice layer is within the projection range of the device's maximum laser filling spot, the initial black and white mask image is sent to the DLP, and the DLP auxiliary node is driven to directly project the complete mask image.

[0032] S432. If the pre-printed pattern of a single slice layer exceeds the projection range of the device's maximum laser filling spot, then multiple sub-mask images are sent to the DLP, and the DLP auxiliary node is driven to project multiple sub-mask images sequentially in a predefined order.

[0033] As a further embodiment of the present invention, S44 includes S441 to S442.

[0034] S441. If the pre-printed pattern of a single slice layer is within the projection range of the device's maximum laser fill spot, then drive the SLM laser node to perform a one-time laser melting based on the projection area of ​​the complete mask image.

[0035] S442. If the current layer projects sub-mask images in a predefined order, the SLM laser node controls the laser to perform scanning melting on the projected sub-regions corresponding to each sub-mask image in the same order as the projection.

[0036] As a further embodiment of the present invention, S45 includes S451 to S453.

[0037] S451. Start the AI ​​visual monitoring node to monitor and capture the molten pool image in real time.

[0038] S452. Analyze the captured molten pool image to obtain the geometric features and temperature distribution data of the molten pool.

[0039] S453. Input the geometric features and temperature distribution data into the ROS2 decision node. The ROS2 decision node dynamically adjusts the laser power or scanning speed based on the analysis results to achieve closed-loop adaptive control.

[0040] Secondly, the present invention provides a surface exposure rapid prototyping device based on ROS2, which includes an initialization module, a slicing module, a rasterization module and a loop module.

[0041] The initialization module is used to drive each hardware node to perform zero initialization based on ROS2, control the molding platform to keep it level with the powder spreading plane, and complete powder replenishment and scraper reset.

[0042] The slicing module is used to import the 3D model of the part to be formed into ROS2-based slicing software, perform posture optimization and generate support structures, and slice according to the preset layer thickness to obtain the 2D contour of each slice layer.

[0043] The rasterization module is used to rasterize the two-dimensional contours of each slice layer to generate the corresponding black and white mask image.

[0044] The loop module is used to repeatedly execute the following submodules for each slice layer until all slice layers are completed:

[0045] The powder supply submodule is used to control the servo motor of the forming chamber to descend according to the preset layer thickness, and to supply powder from the powder supply chamber.

[0046] The powder spreading module is used to spread and level the powder on the forming platform, so that the powder layer is evenly spread.

[0047] The projection submodule is used to send the black and white mask image corresponding to the current slice layer to the DLP and drive the DLP auxiliary node to perform projection.

[0048] The melting submodule is used to drive the SLM laser node to perform selective laser melting according to the projected area.

[0049] The analysis submodule is used to start the AI ​​vision monitoring node to capture and analyze the molten pool image in real time, and input the analysis results to the ROS2 decision node to dynamically adjust the process parameters and achieve closed-loop control.

[0050] Thirdly, the present invention provides a ROS2-based surface exposure rapid prototyping system, comprising a processor, a memory, and a computer program stored in the memory. The computer program can be executed by the processor to implement a ROS2-based surface exposure rapid prototyping method as described in any paragraph of the first aspect.

[0051] Fourthly, the present invention provides a computer-readable storage medium comprising a stored computer program, wherein, when the computer program is executed, it controls the device on which the computer-readable storage medium is located to perform a surface exposure rapid prototyping method based on ROS2 as described in any paragraph of the first aspect.

[0052] By adopting the above technical solution, the present invention can achieve the following technical effects:

[0053] This invention achieves significant efficiency improvements and precision optimizations in additive manufacturing through ROS2-based surface exposure rapid prototyping technology. Employing a ROS2 distributed framework to drive each hardware node enables modular integration and efficient collaboration, reducing system communication latency and ensuring the stability and reliability of the prototyping process. Combining DLP surface exposure and SLM laser melting, it can rapidly manufacture large-sized parts. By using partitioned projection and sequential exposure to process patterns exceeding the equipment's range, manufacturing speed is significantly increased. Simultaneously, an integrated AI vision monitoring node captures real-time images of the molten pool, analyzes geometric features and temperature distribution, and feeds this information back to the ROS2 decision node to dynamically adjust laser power or scanning speed, achieving closed-loop adaptive control. This effectively reduces defects such as porosity and cracks, improving prototyping consistency and quality. Furthermore, automated initialization, intelligent slicing, and rasterization further optimize the entire process, giving the system excellent adaptability and intelligence, making it suitable for rapid prototyping and small-batch production of complex structures, reducing scrap rates and post-processing costs. Attached Figure Description

[0054] To more clearly illustrate the technical solution of the present invention, the accompanying drawings used in the specific embodiments of the present invention will be briefly introduced below. It should be understood that the following drawings only show some specific embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained from these drawings without creative effort.

[0055] Figure 1 This is a schematic diagram of the steps involved in single-layer exposure rapid prototyping.

[0056] Figure 2 This is a schematic diagram of the overall system architecture of ROS2.

[0057] Figure 3 This is a schematic diagram of the system integration framework between ROS2 and OpenVINO.

[0058] Figure 4 This is a schematic diagram of a single-layer pre-printed pattern black and white mask.

[0059] Figure 5 This is a schematic diagram of the division of a single-layer pre-printed pattern area.

[0060] Figure 6 This is a diagram illustrating the preset order of partition projection. Detailed Implementation

[0061] For ease of understanding, the following explains some key terms in this embodiment:

[0062] Robot Operating System 2 (ROS2) is an open-source, distributed framework for building robot applications. In this approach, ROS2 is used to drive and coordinate various hardware nodes, enabling modular integration and efficient communication, thereby ensuring the stable operation of the entire rapid prototyping process.

[0063] Digital Light Processing (DLP) is a projection technique that uses a micromirror array to modulate light. In this method, DLP is used to receive a black-and-white mask image and precisely project it onto a powder layer to define the area for laser melting.

[0064] Selective Laser Melting (SLM) is an additive manufacturing technique that uses a high-energy laser beam to selectively melt metal powder, depositing it layer by layer to form a three-dimensional solid part. In this method, the SLM laser node performs laser melting according to the pattern projected by DLP (Digital Laser Processing), achieving precise part forming.

[0065] A black-and-white mask image is a digital image generated by rasterizing the two-dimensional contour of a part to be formed. White areas typically represent regions that require laser melting, while black areas represent non-melting regions. This image serves as the input for DLP projection.

[0066] An AI vision monitoring node refers to a vision system that integrates artificial intelligence algorithms to capture and analyze images of the molten pool in real time during the forming process. This node can identify the geometric features and temperature distribution of the molten pool, providing data support for subsequent process parameter adjustments.

[0067] The ROS2 decision node refers to the decision module running within the ROS2 framework, responsible for receiving molten pool data analyzed by the AI ​​vision monitoring node. Based on preset control strategies and real-time data, this node dynamically adjusts process parameters such as laser power or scanning speed to achieve closed-loop adaptive control of the molding process.

[0068] The technical solutions of the present invention will now be clearly and completely described with reference to the accompanying drawings in the embodiments of the present invention.

[0069] Example 1, please refer to Figures 1 to 6The first embodiment of the present invention provides a ROS2-based surface exposure rapid prototyping method. It can be executed by a ROS2-based surface exposure rapid prototyping system or device (hereinafter referred to as: surface exposure rapid prototyping system). Specifically, it is executed by one or more processors in the surface exposure rapid prototyping system to implement steps S1 to S4.

[0070] S1. Based on ROS2, each hardware node performs zero initialization, controls the molding platform to keep it level with the powder spreading plane, and completes powder replenishment and scraper reset.

[0071] S2. Import the 3D model of the part to be formed into ROS2-based slicing software, perform posture optimization and generate support structure, and slice according to the preset layer thickness to obtain the 2D contour of each slice layer.

[0072] S3. Rasterize the two-dimensional contours of each slice layer to generate the corresponding black and white mask images.

[0073] S4. Repeat steps S41 to S45 for each slice layer until all slice layers are completed:

[0074] S41. Control the servo motor of the forming chamber to descend according to the preset layer thickness, and supply powder from the powder supply chamber.

[0075] S42. Spread and level the powder on the forming platform to ensure that the powder layer is evenly distributed.

[0076] S43. Send the black and white mask image corresponding to the current slice layer to the DLP and drive the DLP auxiliary node to perform projection.

[0077] S44. Drive the SLM laser node to perform selective laser melting according to the projected area.

[0078] S45. Start the AI ​​visual monitoring node to capture and analyze the molten pool image in real time, and input the analysis results into the ROS2 decision node to dynamically adjust the process parameters and achieve closed-loop control.

[0079] This embodiment employs a surface exposure rapid prototyping control strategy based on the ROS2 distributed framework. The perception layer is used for AI visual inspection, the execution layer is responsible for hardware driving, and the decision layer processes images and performs adaptive control. Preferably, a schematic diagram of the single-layer surface exposure rapid prototyping steps in this embodiment is shown below. Figure 1 As shown.

[0080] The ROS2-based surface exposure rapid prototyping system provided by this invention aims to achieve efficient integration and modular design of equipment through the ROS2 distributed framework, and incorporate AI visual monitoring to perform real-time defect identification and closed-loop feedback control of metal powder layer laying and SLM melting process, thereby reducing communication latency, improving layer thickness uniformity and melt pool stability, which is of great significance for improving the forming accuracy, forming efficiency, process reliability and intelligence level of metal additive manufacturing.

[0081] The numbers S1 to S4 above do not constitute a limitation on the order of the steps, but are merely for ease of reading. Those skilled in the art can adjust the order of the steps according to actual needs. For example, it can be adjusted to steps a to l.

[0082] a. ROS2 driver initializes the hardware node to zero.

[0083] b. ROS2 drives the molding platform to descend to a preset height.

[0084] c. The ROS2 drives the scraper stepper motor node to complete the uniform spreading of powder.

[0085] d. Import the pre-printed part's STL 3D model into ROS2-based slicing software for automatic orientation optimization.

[0086] e. Generate the support structure and prepare for slicing.

[0087] f. Based on the preset layer thickness, slice the 3D model of the pre-printed part to obtain multiple slice layers, and obtain the 2D contour of each slice layer.

[0088] g. Rasterize the two-dimensional contour of each slice layer obtained in step f to generate a black and white mask image.

[0089] h. The mask image obtained in step g is sent to the DLP.

[0090] i.ROS2 drives DLP auxiliary nodes to perform mask image projection of pre-printed parts.

[0091] j.ROS2 drives the SLM laser node to perform selective laser melting.

[0092] k.ROS2 drives AI visual monitoring nodes to achieve closed-loop adaptive control in conjunction with decision nodes.

[0093] l. After a single layer is processed, the servo motor of the forming chamber descends by one layer thickness and enters the next layer cycle. Steps b to c and steps h to k are repeated until all sliced ​​layers are completed, thereby obtaining the required part.

[0094] Specifically, at the start of each molding cycle, the molding chamber can be driven by a stepper motor to descend in preset fixed steps. Simultaneously, the powder supply chamber can be activated, supplying a certain amount of powder material to the molding platform area via gravity or a simple mechanical pusher.

[0095] This method, by introducing the ROS2 distributed framework, achieves efficient collaboration and precise control among hardware nodes, effectively improving the stability and reliability of basic operations such as initialization and powder placement. Combined with DLP surface exposure technology, it can quickly complete the molding of large-sized parts, significantly improving manufacturing efficiency. Simultaneously, the closed-loop control mechanism of AI visual monitoring and ROS2 decision nodes allows for real-time adjustment of process parameters during the melting process, thereby effectively reducing molding defects, improving part precision and consistency, and lowering scrap rates and post-processing costs.

[0096] Based on the above embodiments, in an optional embodiment of the present invention, S1 includes S11 to S13.

[0097] S11 and ROS2 drive the molding chamber servo motor to move to the zero position. If the molding platform and the powder spreading plane are not kept horizontally aligned, control the molding platform to lift up or lower down until it is kept horizontally aligned.

[0098] S12 and ROS2 drive the powder supply servo motor to control the powder supply hopper to descend to the powder filling position to complete powder replenishment, and then rise to the initial powder supply height.

[0099] S13 and ROS2 drive the scraper stepper motor control node, driving the scraper to return to the safe position.

[0100] Figure 2 The overall system architecture of ROS2 is shown. Preferably, this architecture adopts a distributed modular architecture.

[0101] In this embodiment, ROS2 drives each module node to complete the following steps: First, the forming chamber servo motor is controlled to move to the zero position, making the forming platform level with the powder spreading plane. Then, the powder supply chamber servo motor drives the chamber to descend to the powder loading position to complete powder replenishment, and then rises to the initial powder supply height. Finally, the scraper stepper motor drives the scraper to return to the safe position. If the forming platform and the powder spreading plane are not level, the forming platform needs to be raised or lowered to make it flush with the powder spreading plane.

[0102] The ROS2 system communicates with the forming chamber servo motor, sending commands to move it to a preset zero position. This zero position is a precise reference position for the forming platform before forming begins. Simultaneously, to ensure the forming platform maintains strict horizontal alignment with the powder-spreading plane, the system can integrate high-precision sensors, such as laser rangefinders or tilt sensors, to detect the relative height and levelness between the two in real time. If any deviation is detected, the ROS2 decision node calculates the required precise adjustment based on the sensor feedback data and drives the forming chamber servo motor or auxiliary linear actuators to perform fine adjustments, such as controlling the platform's lifting or tilting, until the sensor feedback data indicates that the forming platform and the powder-spreading plane have reached the preset horizontal alignment. This process can be iterative to ensure extremely high alignment accuracy.

[0103] Based on the above embodiments, in an optional embodiment of the present invention, S2 includes S21 to S24.

[0104] S21. Import the STL 3D model of the part to be formed.

[0105] S22. Automatically perform posture optimization on the STL 3D model.

[0106] S23. Generate a support structure for molding based on the optimized model.

[0107] S24. The STL three-dimensional model is sliced ​​based on the preset layer thickness parameters to obtain multiple slice layers, and the two-dimensional contour of each slice layer is obtained.

[0108] Specifically, the STL 3D model of the pre-printed part is imported, the posture optimization is automatically performed, and the necessary support structure is generated. Multiple slices are obtained by slicing according to the preset layer thickness parameters, and the 2D contour of the part in each slice is obtained.

[0109] The imported STL 3D model undergoes automatic pose optimization. Pose optimization aims to adjust the orientation of the 3D model in print space to achieve optimal printing results. This typically includes reducing the required support structure volume, shortening print time, improving the surface quality of the finished part, reducing the risk of warping, and optimizing material utilization. The optimization algorithm integrated within the slicing software comprehensively considers the model's geometric features (such as center of gravity, minimum bounding box, and overhang angle), the physical properties of the printing material, and the process parameters of the printing equipment. For example, the algorithm might iteratively calculate to find a printing orientation that minimizes the model's overhang area or support volume, or to find a pose that maximizes print efficiency.

[0110] Based on the above embodiments, in an optional embodiment of the present invention, S3 includes S31 to S34.

[0111] S31. Rasterize the two-dimensional contours of each slice layer to generate corresponding black-and-white mask images, including:

[0112] S32. Map the two-dimensional contours of each slice layer to pixel coordinates to generate an initial black and white mask image. In the black and white mask image, white pixels represent areas that need to be melted and shaped, and black pixels represent non-molding areas.

[0113] S33. Determine whether the pre-printed pattern of a single slice layer exceeds the projection range of the device's maximum laser filling spot.

[0114] S34. If the projection range is exceeded, the initial black and white mask image is divided into multiple sub-regions in an M×N checkerboard pattern, and corresponding sub-mask images are generated as images to be projected sequentially. Preferably, the M×N checkerboard pattern is a 3×3 pattern, that is, the complete black and white mask pattern is divided into nine equal sub-regions.

[0115] Specifically, the two-dimensional contours of the parts in each slice layer are rasterized and mapped to pixel coordinates to generate a black-and-white mask image. For example... Figure 4 The image shown is a black and white mask image generated by rasterizing the two-dimensional contour of a pre-printed part obtained by slicing a single layer.

[0116] In addition, if the pre-printed pattern of a single slice layer exceeds the projection range of the device's maximum laser filling spot, the complete mask pattern is divided into 9 sub-regions in a 3×3 checkerboard pattern, corresponding sub-mask patterns are generated, and they are exposed sequentially. Figure 5 A schematic diagram showing the division of a single-layer pre-printed pattern area;

[0117] After obtaining the 2D contours of each slice layer, the first step is to convert this continuous geometric information into discrete digital image information. This is achieved by mapping the 2D contours of each slice layer to pixel coordinates, thereby generating an initial black-and-white mask image. This mapping process typically involves image processing algorithms; for example, for each slice layer's 2D contour, the system determines its minimum bounding rectangle on the forming plane and maps it to a pixel grid of a preset resolution. Pixels inside the contour are marked as "white" or "1," indicating that the area needs exposure or melting; while pixels outside the contour are marked as "black" or "0," indicating that no exposure or melting is required. This process ensures that the 2D contours can be recognized and projected by the DLP device and work collaboratively with the SLM laser node.

[0118] The system then determines whether the pre-printed pattern of a single slice layer exceeds the projection range of the device's maximum laser fill spot. This determination is a crucial step, designed to assess whether the current slice layer can be completed with a single projection and melting operation. The system calculates the dimensions of the initial black-and-white mask image (e.g., pixel width and height) and compares it to the maximum projection size preset by the DLP device or the effective projection area size of the maximum laser fill spot of the SLM device. If any dimension of the mask image (width or height) exceeds the corresponding maximum projection range of the device, the pattern is considered to be outside the projection range. This determination can be performed automatically in the slicing software or by the ROS2 decision node before image transmission.

[0119] If the result indicates that the projection range is exceeded, further processing of the initial black-and-white mask image is required. In this case, the system divides the initial black-and-white mask image into multiple sub-regions using an M×N checkerboard pattern and generates corresponding sub-mask images as images to be projected sequentially. This division method means dividing the original initial black-and-white mask image into M columns horizontally and N rows vertically, forming M×N rectangular sub-regions. The size of each sub-region should be less than or equal to the maximum projection range of the device to ensure that each sub-mask image can be effectively projected by the DLP device. During the division process, a certain overlap of pixel areas can be considered between adjacent sub-regions to ensure the continuity of the subsequent molten area and avoid gaps. For each divided sub-region, the system extracts the pixel data of the corresponding part of the initial black-and-white mask image to generate an independent sub-mask image. These sub-mask images are stored and subsequently projected in a predefined order (e.g., from left to right, from top to bottom), thereby gradually completing the formation of the entire slice layer.

[0120] Through the above technical solution, when the size of the part to be formed exceeds the maximum projection range of the DLP equipment or the maximum laser filling spot projection range of the SLM equipment, it is no longer limited by the area restriction of single projection or melting. By intelligently dividing the initial black and white mask image into multiple sub-regions and generating corresponding sub-mask images, each sub-region can be accurately projected and laser melted within the effective working range of the equipment. This not only effectively solves the problem of insufficient projection range of the equipment when forming large-sized parts, avoiding the dilemma of being unable to form due to size limitations or requiring multiple complex mechanical movements, but also ensures the continuity and forming accuracy of the entire slice layer through sequential projection and melting. This solution improves the applicability and flexibility of the equipment, enabling the ROS2-based surface exposure rapid prototyping method to be applied to a wider range of part sizes, thereby improving overall production efficiency and forming quality.

[0121] S41 specifically refers to: ROS2 controlling the servo motor of the molding chamber to precisely descend according to the preset layer thickness. Powder is supplied from the powder supply chamber to the molding platform area.

[0122] ROS2 enables high-precision, real-time closed-loop control of the descent process of the molding chamber servo motor, ensuring that the descent distance of each layer precisely matches the preset layer thickness, thereby effectively controlling the vertical dimensional accuracy of the molded parts. Simultaneously, under the coordination of ROS2, the powder supply chamber can provide powder to the molding platform area in a timely and appropriate manner, ensuring sufficient and uniform powder for subsequent powder layers. This precise descent control and coordinated powder supply mechanism guarantee the quality of the powder layer from the source, avoiding molding defects caused by uneven layer thickness or insufficient powder supply. This significantly improves the dimensional accuracy, surface quality, and mechanical properties of the final molded parts, providing a solid foundation for achieving high-quality, high-precision rapid molding.

[0123] S42 specifically refers to the ROS2 drive scraper stepper motor control node. The scraper stepper motor drives the scraper to perform reciprocating motion to achieve uniform powder spreading.

[0124] This application addresses the potential uniformity issues during powder deposition by introducing a ROS2-driven stepper motor control node to precisely control the stepper motor and drive the squeegee in reciprocating motion. This refined control allows the squeegee to operate stably at a preset speed and trajectory, ensuring that each layer of powder is uniformly and evenly deposited on the forming platform. Compared to powder scraping methods lacking precise control, this approach significantly improves the uniformity and density of the powder layer, providing a stable foundation for the subsequent laser melting process and effectively avoiding melting defects, part warping, or decreased mechanical properties caused by uneven powder layers. Ultimately, by achieving highly uniform powder deposition, this application significantly improves the precision, surface quality, and overall performance of the formed parts, providing a more reliable and efficient powder processing step for ROS2-based surface exposure rapid prototyping methods.

[0125] Based on the above embodiments, in an optional embodiment of the present invention, S43 includes S431 to S432.

[0126] S431. If the pre-printed pattern of a single slice layer is within the projection range of the device's maximum laser filling spot, the initial black and white mask image is sent to the DLP, and the DLP auxiliary node is driven to directly project the complete mask image.

[0127] S432. If the pre-printed pattern of a single slice layer exceeds the projection range of the device's maximum laser filling spot, then multiple sub-mask images are sent to the DLP, and the DLP auxiliary node is driven to project the multiple sub-mask images sequentially in a predefined order. The predefined order is an "S"-shaped scanning sequence from left to right and from top to bottom within the 3×3 checkerboard grid.

[0128] Specifically, the mask image obtained in step S3 is sent to the DLP, and then ROS2 drives the DLP auxiliary node to perform mask image projection of the pre-printed part. If the pre-printed pattern of a single slice layer is within the projection range of the device's maximum laser filling spot, there is no need to divide the area, and it can be projected directly. Figure 4 The complete mask image is shown. If the pre-printed pattern of a single slice layer exceeds the projection range of the device's maximum laser fill spot, then according to... Figure 5 Divide the area, according to Figure 6 The predefined order shown projects multiple sub-mask images sequentially.

[0129] This application effectively addresses the issue of pre-printed patterns on a single slice layer exceeding the maximum laser filling spot projection range of the DLP equipment during the molding of large-sized parts. When the pattern is within the projection range, the system directly projects the complete mask image, ensuring molding efficiency. When the pattern exceeds the projection range, the system divides the pattern into multiple sub-mask images and projects them sequentially in a predefined order, ensuring complete coverage and accurate molding of the entire large-sized pattern. This segmented, sequential projection method not only overcomes the limitations of traditional surface exposure technology in processing large-sized parts, avoiding molding defects caused by insufficient projection range, but also ensures the accuracy and continuity of the projection process through the coordinated control of ROS2, thereby improving the molding quality and applicability of large-sized parts.

[0130] Based on the above embodiments, in an optional embodiment of the present invention, S44 includes S441 to S442.

[0131] S441. If the pre-printed pattern of a single slice layer is within the projection range of the device's maximum laser fill spot, then drive the SLM laser node to perform a one-time laser melting based on the projection area of ​​the complete mask image.

[0132] S442. If the current layer projects sub-mask images in a predefined order, the SLM laser node controls the laser to perform scanning melting on the projected sub-regions corresponding to each sub-mask image in the same order as the projection.

[0133] Specifically, ROS2 drives the SLM laser node to perform selective laser melting. If the pre-printed pattern of a single slice layer is within the projection range of the device's maximum laser fill spot, no area division is required, and the process can directly proceed to step S44 to perform laser projection melting. If the pre-printed pattern of a single slice layer exceeds the projection range of the device's maximum laser fill spot, then... Figure 6 The laser projection melting is performed sequentially according to the predefined order shown.

[0134] This application achieves precise coordination between the SLM laser melting process and the DLP projection process. When the slice pattern is small and within the range of a single projection, the laser can melt the entire area in one go, improving efficiency. However, when the slice pattern is large and requires regional projection of sub-mask images, the SLM laser node can sequentially perform scanning melting on each projected sub-region according to the same predefined order as DLP projection. This synchronous melting mechanism effectively avoids misalignment, overlap, or omissions caused by asynchronous projection and melting, ensuring the accuracy and continuity of the melted area. This significantly improves the forming accuracy and surface quality of large or complex parts, while optimizing overall forming efficiency.

[0135] Based on the above embodiments, in an optional embodiment of the present invention, S45 includes S451 to S453.

[0136] S451. Start the AI ​​visual monitoring node to monitor and capture the molten pool image in real time.

[0137] S452. Analyze the captured molten pool image to obtain the geometric features and temperature distribution data of the molten pool.

[0138] S453. Input the geometric features and temperature distribution data into the ROS2 decision node. The ROS2 decision node dynamically adjusts the laser power or scanning speed based on the analysis results to achieve closed-loop adaptive control.

[0139] Specifically, ROS2 activates the AI ​​vision monitoring node to monitor and capture molten pool images in real time, analyze the geometric features and temperature distribution of the molten pool, and input the analysis results to the ROS2 decision node. Based on the analysis results, the process parameters are dynamically adjusted to achieve closed-loop adaptive control.

[0140] In the forming cycle of each slice layer, the system first completes processes such as powder supply, powder spreading, projection, and selective laser melting. When entering the melting stage, the analysis process is further initiated: the melting process is perceived online through AI visual monitoring nodes, and "monitoring-analysis-feedback parameter adjustment" is embedded into the cycle steps of that layer, thereby upgrading the melting process from "open-loop execution" to "closed-loop execution with feedback".

[0141] At the hardware level, the system can construct a coaxial or paraxial precision visual monitoring optical path, integrating a high-speed industrial camera and an infrared thermal imager. To ensure imaging quality, a filter group for specific wavelengths is added to the front end of the sensor to effectively filter out interference light reflected from high-energy lasers, thereby clearly capturing the visible light geometry and infrared thermal radiation signals of the molten pool in real time. These sensors are connected to the edge computing unit running the ROS2 system via a high-speed data interface, providing computing power for low-latency processing of massive image data.

[0142] At the software and algorithm level, the system incorporates a dedicated AI visual analysis node. This node first dynamically crops the region of interest (ROI) centered on the molten pool from the original image based on feedback from the galvanometer scanning position, reducing data redundancy from irrelevant backgrounds. Subsequently, through image enhancement and adaptive threshold segmentation algorithms, the contour of the molten pool is accurately extracted, and key geometric features such as its width, length, area, and aspect ratio are calculated. Simultaneously, infrared image data is combined to map the temperature field distribution inside the molten pool, obtaining the highest temperature point and the direction of the temperature gradient. For complex molten pool morphologies, the system can further utilize a lightweight deep learning model to identify online abnormalities such as spheroidization, porosity, or excessive spatter.

[0143] Based on ROS2's efficient communication mechanism, the analysis node encapsulates the calculated molten pool characteristic parameters into specific message types and publishes them in real time through ROS2 topics. After subscribing to the data in this topic, the decision control node compares the measured molten pool state (such as current width and average temperature) with preset standard process thresholds. Using a built-in incremental PID control algorithm or fuzzy control logic, the system calculates the compensation amount for laser power or scanning speed in real time based on the deviation value. For example, when a sharp decrease in molten pool area and low temperature are detected, the algorithm determines that the energy is insufficient and generates a command to increase the laser power.

[0144] Ultimately, the generated adjustment commands are sent to the underlying hardware driver node via the ROS2 service interface, enabling millisecond-level dynamic adjustment of process parameters. This closed-loop control mechanism allows the system to instantly correct processing parameters based on the actual thermal state of the molten pool during each layer's scanning process. This effectively suppresses defects such as overheating and incomplete fusion caused by heat accumulation or uneven powder distribution, significantly improving the stability of the additive manufacturing process and the density of the formed parts.

[0145] The AI ​​visual monitoring node is a module integrating image acquisition hardware and image processing software. Its core function is to acquire real-time, continuous image data of the laser melting area. In practice, one or more high-speed industrial cameras can be installed inside the SLM equipment, such as above or to the side of the forming chamber. These cameras can include visible light cameras to capture the geometry and size of the molten pool, or infrared thermal imagers to acquire temperature distribution information. The AI ​​visual monitoring node interacts with the entire system via a ROS2 communication interface. Upon receiving the signal indicating the start of laser melting, it continuously captures images of the molten pool area at a preset high frame rate (e.g., hundreds to thousands of frames per second). The captured image data is accompanied by precise timestamps for subsequent synchronous analysis with current process parameters.

[0146] This step aims to transform the raw image data captured by the AI ​​visual monitoring node into quantifiable, physically meaningful process parameters. For visible light images, a series of image processing algorithms, such as edge detection, thresholding, and morphological operations, can be used to accurately identify the boundaries of the molten pool and further calculate geometric feature parameters such as the length, width, area, and shape factor of the molten pool. For example, by setting an appropriate grayscale threshold, the molten pool area can be effectively separated from the background, and then its precise dimensions can be calculated using connected component analysis. For images captured by infrared thermal imagers, temperature distribution data of the molten pool surface can be directly extracted. This typically involves calibrating the grayscale values ​​of image pixels and converting them into actual temperature values. Based on this, key data such as the highest temperature, average temperature, and temperature gradient of the molten pool can be further analyzed. In addition, by combining deep learning models, such as convolutional neural networks (CNNs), the model can be trained to automatically identify abnormal morphologies of the molten pool, such as spatter, voids, or discontinuous melt channels, as well as uneven temperature distribution, thereby extracting key features more robustly and intelligently.

[0147] The ROS2 decision node is the core "brain" of the entire closed-loop control system. It receives real-time feedback data from the AI ​​visual monitoring node and calculates the optimal process parameter adjustments based on preset control strategies or machine learning-based adaptive models. Subsequently, this node sends the adjustment commands to the SLM laser node via the ROS2 communication mechanism. Specifically, the ROS2 decision node can integrate PID controllers, fuzzy logic controllers, or more complex adaptive control modules based on artificial intelligence algorithms such as reinforcement learning. For example, when analysis results indicate that the melt pool width is too large or the temperature is too high, the decision node will instruct the SLM laser node to reduce laser power or increase scanning speed; conversely, if the melt pool is too narrow or the temperature is too low, it may instruct to increase laser power or decrease scanning speed. These control strategies can be based on pre-established process windows, physical models, or AI models trained with extensive experimental data. The ROS2 decision node sends the adjusted laser power and scanning speed commands to the SLM laser node in real time through the ROS2 communication mechanism. After receiving the commands, the SLM laser node immediately adjusts the output power of the laser and the scanning speed of the galvanometer, thereby realizing real-time, closed-loop, adaptive control of the melting process and ensuring that the molten pool state is always maintained within the optimal process window.

[0148] Through the above technical solution, this application introduces a real-time AI visual monitoring and closed-loop adaptive control mechanism into the ROS2-based surface exposure rapid prototyping method. This effectively solves the technical problem of unstable molding quality and easy defects caused by the lack of real-time feedback and dynamic adjustment in the melting process of traditional SLM processes. Specifically, the AI ​​visual monitoring node captures and analyzes the geometric features and temperature distribution data of the molten pool in real time, which directly reflects the actual state of the melting process. The ROS2 decision node uses this real-time and accurate feedback data to quickly determine whether the current melting state deviates from the ideal process window. Once a deviation is detected, the decision node immediately calculates the optimal laser power or scanning speed adjustment and sends the instruction to the SLM laser node through the ROS2 communication mechanism, thereby correcting the melting parameters in a very short time. This dynamic adjustment capability transforms the entire molding process from static preset parameter control to dynamic real-time optimization control, significantly improving the quality stability of the molded parts, reducing the defect rate, and better adapting to process fluctuations caused by different materials, complex geometries, and environmental changes. Especially in scenarios involving regional projection and melting in the aforementioned methods, AI visual monitoring can independently monitor and adjust the melting status of each sub-region, further improving the forming accuracy and quality of local areas. Ultimately, this application achieves higher precision, higher quality, and fewer defects in part manufacturing, reduces reliance on operator experience, and improves production efficiency and material utilization.

[0149] Example 2: The present invention provides a surface exposure rapid prototyping device based on ROS2, which includes an initialization module, a slicing module, a rasterization module and a loop module.

[0150] The initialization module is used to drive each hardware node to perform zero initialization based on ROS2, control the molding platform to keep it level with the powder spreading plane, and complete powder replenishment and scraper reset.

[0151] The slicing module is used to import the 3D model of the part to be formed into ROS2-based slicing software, perform posture optimization and generate support structures, and slice according to the preset layer thickness to obtain the 2D contour of each slice layer.

[0152] The rasterization module is used to rasterize the two-dimensional contours of each slice layer to generate the corresponding black and white mask image.

[0153] The loop module is used to repeatedly execute the following submodules for each slice layer until all slice layers are completed:

[0154] The powder supply submodule is used to control the servo motor of the forming chamber to descend according to the preset layer thickness, and to supply powder from the powder supply chamber.

[0155] The powder spreading module is used to spread and level the powder on the forming platform, so that the powder layer is evenly spread.

[0156] The projection submodule is used to send the black and white mask image corresponding to the current slice layer to the DLP and drive the DLP auxiliary node to perform projection.

[0157] The melting submodule is used to drive the SLM laser node to perform selective laser melting according to the projected area.

[0158] The analysis submodule is used to start the AI ​​vision monitoring node to capture and analyze the molten pool image in real time, and input the analysis results to the ROS2 decision node to dynamically adjust the process parameters and achieve closed-loop control.

[0159] Example 3: This invention provides a ROS2-based surface exposure rapid prototyping system, which includes a processor, a memory, and a computer program stored in the memory. The computer program can be executed by the processor to implement a ROS2-based surface exposure rapid prototyping method as described in any paragraph of Example 1.

[0160] It is understood that the ROS2-based surface exposure rapid prototyping system can be a ROS2 robot platform or a surface exposure 3D printer (photopolymerization 3D printer).

[0161] Example 4: The present invention provides a computer-readable storage medium, the computer-readable storage medium including a stored computer program, wherein, when the computer program is executed, it controls the device where the computer-readable storage medium is located to execute a ROS2-based rapid prototyping method as described in any paragraph of Example 1.

[0162] Obviously, the embodiments described above are only some embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0163] In the several embodiments provided in this invention, it should be understood that the disclosed apparatus and methods can also be implemented in other ways. The apparatus and method embodiments described above are merely illustrative. For example, the flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of apparatus, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than those marked in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram and / or flowchart, and combinations of blocks in block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0164] In addition, the functional modules in the various embodiments of the present invention can be integrated together to form an independent part, or each module can exist independently, or two or more modules can be integrated to form an independent part.

[0165] If the aforementioned functions are implemented as software functional modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, electronic device, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory, random access memory, magnetic disks, or optical disks. It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0166] The terminology used in the embodiments of this invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The singular forms “a,” “the,” and “the” used in the embodiments of this invention are also intended to include the plural forms unless the context clearly indicates otherwise.

[0167] It should be understood that the term "and / or" used in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. Additionally, the character " / " in this article generally indicates that the preceding and following related objects have an "or" relationship.

[0168] Depending on the context, the word "if" as used here can be interpreted as "when," "when," "in response to determination," or "in response to detection." Similarly, depending on the context, the phrase "if determination" or "if detection (of the stated condition or event)" can be interpreted as "when determination," "in response to determination," "when detection (of the stated condition or event)," or "in response to detection (of the stated condition or event)."

[0169] The terms "first" and "second" used in the embodiments are merely to distinguish similar objects and do not represent a specific ordering of objects. It is understood that "first" and "second" can be interchanged in a specific order or sequence where permitted. It should be understood that the objects distinguished by "first" and "second" can be interchanged where appropriate so that the embodiments described herein can be implemented in an order other than those illustrated or described herein.

[0170] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A rapid prototyping method based on ROS2 surface exposure, characterized in that, Include: Based on ROS2, each hardware node performs zero initialization, controls the molding platform to keep it level with the powder spreading plane, and completes powder replenishment and scraper reset. Import the 3D model of the part to be formed into ROS2-based slicing software, perform posture optimization and generate support structure, and slice according to preset layer thickness to obtain the 2D contour of each slice layer. The two-dimensional contours of each slice layer are rasterized to generate corresponding black and white mask images; Repeat the following steps for each slice layer until all slice layers are completed: The servo motor of the forming chamber is controlled to descend according to the preset layer thickness, and powder is supplied by the powder supply chamber; The powder is spread and leveled on the forming platform to ensure that the powder layer is evenly laid. Send the black and white mask image corresponding to the current slice layer to the DLP and drive the DLP auxiliary node to perform projection; Drive the SLM laser node to perform selective laser melting based on the projected area; The AI ​​visual monitoring node is activated to capture and analyze images of the molten pool in real time. The analysis results are then input into the ROS2 decision node to dynamically adjust process parameters and achieve closed-loop control.

2. The method for rapid prototyping based on ROS2 surface exposure according to claim 1, characterized in that, Based on ROS2, each hardware node performs a zero-initialization, specifically including: ROS2 drives the servo motor of the forming chamber to move to the zero position. If the forming platform and the powder spreading plane are not kept horizontally aligned, the forming platform is controlled to be raised or lowered until it is kept horizontally aligned. ROS2 drives the powder supply servo motor to control the powder supply hopper to descend to the powder filling position to complete powder replenishment, and then rises to the initial powder supply height; ROS2 drives the scraper stepper motor control node, driving the scraper to return to a safe position.

3. The method for rapid prototyping based on ROS2 surface exposure according to claim 1, characterized in that, Importing the 3D model into slicing software and slicing it to obtain the 2D contour includes: Import the STL 3D model of the part to be formed; Automatic pose optimization is performed on the STL 3D model; The optimized model is used to generate the support structure for molding; The STL 3D model is sliced ​​based on preset layer thickness parameters to obtain multiple slice layers, and the 2D contour of each slice layer is obtained.

4. The method for rapid prototyping based on ROS2 surface exposure according to claim 1, characterized in that, The two-dimensional contours of each slice layer are rasterized to generate corresponding black-and-white mask images, including: The two-dimensional contours of each slice layer are mapped to pixel coordinates to generate an initial black and white mask image; Determine whether the pre-printed pattern of a single slice layer exceeds the projection range of the device's maximum laser fill spot; If the projection range is exceeded, the initial black and white mask image is divided into multiple sub-regions in an M×N checkerboard pattern, and multiple corresponding sub-mask images are generated as images to be projected in sequence.

5. The method for rapid prototyping based on ROS2 surface exposure according to claim 1, characterized in that, The black-and-white mask image corresponding to the current slice layer is sent to the DLP, and the DLP auxiliary node is driven to perform projection, specifically including: If the pre-printed pattern of a single slice layer is within the projection range of the device's maximum laser fill spot, the initial black and white mask image is sent to the DLP, and the DLP auxiliary node is driven to directly project the complete mask image. If the pre-printed pattern of a single slice layer exceeds the projection range of the device's maximum laser filling spot, multiple sub-mask images are sent to the DLP, and the DLP auxiliary node is driven to project the multiple sub-mask images sequentially in a predefined order.

6. The method for rapid prototyping based on ROS2 surface exposure according to claim 5, characterized in that, Driving the SLM laser node to perform selective laser melting based on the projected area includes: If the pre-printed pattern of a single slice layer is within the projection range of the device's maximum laser fill spot, the SLM laser node is driven to perform a one-time laser melting based on the projection area of ​​the complete mask image; If the current layer projects sub-mask images in a predefined order, the SLM laser node controls the laser to perform scanning melting on the projected sub-regions corresponding to each sub-mask image in the same order as the projection.

7. The method for rapid prototyping based on ROS2 surface exposure according to claim 6, characterized in that, The AI ​​visual monitoring node is activated to capture and analyze molten pool images in real time, specifically including: Activate the AI ​​visual monitoring node to monitor and capture molten pool images in real time; The captured images of the molten pool are analyzed to obtain the geometric features and temperature distribution data of the molten pool. The geometric features and temperature distribution data are input into the ROS2 decision node, which dynamically adjusts the laser power or scanning speed based on the analysis results to achieve closed-loop adaptive control.

8. A surface exposure rapid prototyping apparatus based on ROS2, characterized in that, Include: The initialization module is used to drive each hardware node to perform zero initialization based on ROS2, control the molding platform to keep it level with the powder spreading plane, and complete powder replenishment and scraper reset. The slicing module is used to import the 3D model of the part to be formed into the ROS2-based slicing software, perform posture optimization and generate support structure, and slice according to the preset layer thickness to obtain the 2D contour of each slice layer. The rasterization module is used to rasterize the two-dimensional contours of each slice layer to generate the corresponding black and white mask images. The loop module is used to repeatedly execute the following submodules for each slice layer until all slice layers are completed: The powder supply submodule is used to control the servo motor of the forming chamber to descend according to the preset layer thickness, and to supply powder from the powder supply chamber. The powder spreading module is used to spread and level the powder on the forming platform, so that the powder layer is evenly spread. The projection submodule is used to send the black and white mask image corresponding to the current slice layer to the DLP and drive the DLP auxiliary node to perform projection. The melting submodule is used to drive the SLM laser node to perform selective laser melting according to the projected area; The analysis submodule is used to start the AI ​​vision monitoring node to capture and analyze the molten pool image in real time, and input the analysis results to the ROS2 decision node to dynamically adjust the process parameters and achieve closed-loop control.

9. A surface exposure rapid prototyping system based on ROS2, characterized in that, It includes a processor, a memory, and a computer program stored in the memory; the computer program can be executed by the processor to implement a ROS2-based rapid prototyping method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored computer program, wherein, when the computer program is executed, it controls the device on which the computer-readable storage medium is located to perform a surface exposure rapid prototyping method based on any one of claims 1 to 7.

Images