Reconstructing editable prismatic CAD from rounded voxel models

In this paper, a systematic scheme is proposed and novel technologies are developed to automatically reconstruct a CAD model from a set of point clouds scanned from the boundary surface of an existing object. The proposed scheme is composed of three major steps. In the first step, multiple input point clouds are incrementally integrated into a watertight triangle mesh to recover the object shape. In the second step, mesh segmentation is applied to the triangle mesh to extract individual geometric feature surfaces. Finally, the manifold topology describing the connectivity information between different geometric surfaces is automatically extracted and the mathematical description of each geometric feature is computed. The computed topology and geometry information represented in ACIS modeling kernel form a CAD model that may be used for various downstream applications. Compared with prior work, the proposed approach has the unique advantage that the processes of recognizing geometric features and of reconstructing CAD models are fully automated. Integrated with state of the art scanning devices, the developed model reconstruction method can be used to support reverse engineering of high precision mechanical components. It has potential applications to many engineering problems with a major impact on rapid design and prototyping, shape analysis, and virtual reality.

Optimizing CAD Models with Latent Space Manipulation

Internet technology opens up another domain for building future CAD/CAM environment. The environment will be global, network-centric, and spatially distributed. In this paper, we present a new approach to network-centric virtual prototyping (NetVP) in a distributed design environment. The presented approach combines the current virtual assembly modeling and analysis technique with distributed computing and communication technology for supporting virtual prototyping activities over the network. This paper focuses on interoperability, shape representation, and geometric processing for distributed virtual prototyping. STEP standard and CORBA-based interfaces allow the bi-directional communication between the CAD model and virtual prototyping model, which makes it possible to solve the problems of interoperability, heterogeneity of platforms, and data sharing. STEP AP203 is utilized as a means of transferring and sharing product models. In addition, Attributed Abstracted B-rep (AAB) is introduced as 3D shape abstraction for transparent and efficient transmission of 3D models and for the maintenance of naming consistency between CAD models and virtual prototyping models over the network. Further, this paper discusses geometric processings needed for distributed virtual prototyping activities such as collision detection and interactive assembly modeling.

The article describes methods and techniques intended for designing and manufacturing scaled airplane models applicable in experimental flight testing. Reduced model with resized geometric dimensions should be constructed in such way to preserve similarity of its structural properties with properties of the real aircraft. Having kept the similarity of weight, stiffness and aerodynamic load distribution and maintaining properly scaled values of thrust-to-weight and wing loading ratios we can prepare scaled model showing in-flight features identical to the real airplane.A series of measuring and technological activities have to be undertaken to develop full CAD model mapping airframe geometry and structure of the real aircraft. The process chain includes measuring, design and technological activities. The development of the geometry model includes: scanning outer geometry of the real airframe, development of the surface geometry model in the CAD environment and then introducing inner geometry items which together with outer skin face make a completely assembled model of the aircraft. For the needs of constructing scaled model the full-size CAD model has to be reduced. The structure of that resized model do not have to be the same as the real one. Materials and structural elements can be quite different but after assembling them the similarity of static and dynamic structural effects should be proved. Within the manufacturing phase there are some necessary stages like preparing molds for laminating components, milling solid elements, assembly of structural components, surface finishing of the model-body surface. The final model was subjected to a load test in a special test stand to check its deformability under specific load. In the same way the numerical static analysis of the full-size FEM model was conducted. The distribution of static displacement results obtained in both cases were compared. The other way of checking structural properties of the scaled model is comparable analysis of resonant vibrations. However, in this case, vibration measurements were not carried out and model verification stopped on a comparative analysis of construction mass distribution.The procedure and results were elaborated on the example of the Tu-154M aircraft, of which the reduced model was developed in the Faculty of Mechatronics, Armament and Aerospace of the Military University of Technology (FMA MUT Warsaw, PL). Constructing and testing dynamically scaled model is one of many tasks solved in the framework of complex research project, of which the general purpose was to give numerical and experimental results useful in explaining the circumstances of the Smolensk disaster.

Reverse engineering applications for recovery of broken or worn parts and re-manufacturing: Three case studies

Truncated signed distance function (TSDF) based volumetric surface reconstructions of static environments can be readily acquired using recent RGB-D camera based mapping systems. If objects in the environment move then a previously obtained TSDF reconstruction is no longer current. Handling this problem requires segmenting moving objects from the reconstruction. To this end, we present a novel solution to the motion segmentation of TSDF volumes. The segmentation problem is cast as CRF-based MAP inference in the voxel space. We propose: a novel data term by solving sparse multi-body motion segmentation and computing likelihoods for each motion label in the RGB-D image space, and, a novel pair wise term based on gradients of the TSDF volume. Experimental evaluation shows that the proposed approach achieves successful segmentations on reconstructions acquired with Kinect Fusion. Unlike the existing solutions which only work if the objects move completely from their initially occupied spaces, the proposed method permits segmentation of objects when they start to move.

The plant and shipbuilding industry is different from other industries owing to its complex layout, uniquely designed products, and a long life cycle of products. Client organisations are endeavouring to use the information obtained from the design during the operation and maintenance period; therefore, they require the CAD model data from the contractor. When a client requests a CAD model format other than the CAD system possessed by the EPCI company (or contractor), the CAD model format must be converted to the corresponding system. Manual model rebuilding is the primarily used plant and ship CAD model-conversion method; it requires a significant amount of time and resources and converted-model verification. To reduce the costs involved, various studies are being conducted that use the standard and commercial formats. In this study, we analyse and implement the method of converting the hull structure model to another CAD model.

As one of Indonesian city that emerged from the industrial sector, Balikpapan grew and was inhabited by transmigrates from various cultures. The development and growth patterns of the Balikpapan urban area was influenced by the culture of the residents, due to the absence of empire influence or colonialism. This research was conducted to identify the oldest area as a starting point of the urban development patterns of Balikpapan. The building-age profile method was used to identify the oldest building groups or blocks, which belong to the city’s heritage zone. In this analysis, a series of figure-ground images in several periods were collected to indicate the building age. Through the process of overlaying solid-void maps, the pattern of space-built interaction was generated as the patterns of urban development. Results shown that urban development, which was determined through overlay technique, was organic pattern, which was formed by several connected irregular shapes. The conservation of the urban pattern indicates the effort to achieve urban identity, as one of the sustainable urban form indicator.

Accurate air flow measurements have been a highly sought-after discipline in the industry in recent years, especially among manufacturers of HVAC (Heating, Ventilation, and Air Conditioning) systems. One of the challenges for scientific research is the area of simulation and verification of results for sub-components of these systems, such as reducers, diffusers, or control dampers. Although these parts have a significant impact on measurements, they are often greatly simplified and or completely ignored in CAD models by HVAC manufacturers. As a result of this simplification, significant differences arise between laboratory measured results and CFD simulation results. However, to obtain realistic values from the simulation, it is necessary to use 3D data as close to reality as possible. This paper describes the process of creating a realistic model of a control damper that is part of the distribution box of the air ventilation system. This damper was 3D scanned using an ATOS II Triple Scan 5M and processed using GOM Professional. The dimensions of the duct and the angles between the end positions of the control damper were measured. Using reverse engineering in Geomagic Design X and the measured values, a sufficiently accurate 3D model of the control damper was created. The output data from reverse engineering was used as input data for CFD analysis. The CFD simulation of the control damper was performed for 3 different flow rates (10, 20, and 35 m <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">3</sup> /h). Comparisons of simulation results between CAD data from the HVAC manufacturer and data from 3D scanning achieve a difference of up to 35 % (when the control damper is fully closed or partially open). Using a realistic 3D model of the control damper obtained by reverse engineering, the results and outputs of the simulation were significantly refined. The results of this work can be used in the design of distribution boxes or entire HVAC systems.

Improving precision of material extrusion 3D printing by in-situ monitoring &amp; predicting 3D geometric deviation using conditional adversarial networks

Implicit neural representations (INRs) have emerged as a promising framework for representing signals in low-dimensional spaces. This survey reviews the existing literature on the specialized INR problem of approximating signed distance functions (SDFs) for surface scenes, using either oriented point clouds or a set of posed images. We refer to neural SDFs that incorporate differential geometry tools, such as normals and curvatures, in their loss functions as geometric INRs. The key idea behind this 3D reconstruction approach is to include additional regularization terms in the loss function, ensuring that the INR satisfies certain global properties that the function should hold — such as having unit gradient in the case of SDFs. We explore key methodological components, including the definition of INR, the construction of geometric loss functions, and sampling schemes from a differential geometry perspective. Our review highlights the significant advancements enabled by geometric INRs in surface reconstruction from oriented point clouds and posed images.

Named entity recognition serves as a cornerstone in natural language processing and has garnered extensive research attention due to its significance in various downstream applications. Owing to the intricate nature of Chinese texts, characterized by complex syntactic structures and the lack of explicit word boundaries, conventional NER methodologies often encounter difficulties in simultaneously optimizing recognition accuracy and computational efficiency. To address this issue, the study proposes a named entity recognition algorithm that integrates attention mechanisms with Convolutional Neural Networks, incorporates into a Transformer-based bidirectional encoder framework for training. A multi-head self-attention mechanism is employed to capture the global semantic information of the text, and multi-task learning is introduced to construct the final model. When evaluated on datasets with sample sizes of 200, 1000, and 3000, the proposed model consistently outperforms the baseline models in terms of precision, recall, and F1 score. Specifically, under the low-resource setting with 200 samples, the model achieves a precision of 98.62%, a recall of 98.10%, and an F1 score of 98.36%. In terms of inference efficiency, the model processes at a speed of 2618 tokens per second. The experimental results indicate that this method can be widely applied in various fields such as information extraction and text understanding, providing strong technical support for related research.

Digital Mockups (Virtual Prototypes) significantly reduce risk and cost on large aerospace programs, but their accuracy hinges on the creation of 3D models of existing facilities, based on "as-built" field conditions instead of less-accurate design blueprints. Methods for creating as-built models vary from tedious hand-measurements, to photograph-rectification-survey techniques (collectively known as photogrammetry), to the evolving technology of terrestrial laser scanning. Lockheed Martin has proven through robust field testing that the accuracy of each technology is variable to specific applications. Therefore, we have developed a hybrid method that leverages the strengths of each technology, including edge recognition, targeting accuracy, and complex-surface collection. This hybridization has helped to identify key technical deficiencies in the current state-of- practice for as-built CAD modeling, such as automatic modeling and feature extraction. In addition, Lockheed Martin has refined a statistical evaluation technique to objectively measure the accuracy of the final model against the field conditions, regardless of the technology utilized. The advantages and disadvantages of each technology are presented here, as well as methodologies for effectively combining the technologies into a more-accurate and efficient hybrid process. This overall capability is crucial to Lockheed Martin for reducing reliance on physical prototypes, and optimizing project return on cost, schedule, and risk.

This paper examines the problem of reconstructing a voxelized representation of 3D space from a series of images. An iterative algorithm is used to find the scene model which jointly explains all the observed images by determining which region of space is responsible for each of the observations. The current approach formulates the problem as one of optimization over estimates of these responsibilities. The process converges to a distribution of responsibility which accurately reflects the constraints provided by the observations, the positions and shape of both solid and transparent objects, and the uncertainty which remains. Reconstruction is robust, and gracefully represents regions of space in which there is little certainty about the exact structure due to limited, non-existent, or contradicting data. Rendered images of voxel spaces recovered from synthetic and real observation images are shown.

The volume and complexity of Biomedical Imaging (BMI) data can be handled by well-known Product Lifecycle Management (PLM) solutions if a research study in this field is modeled as a cyclic process of four phases: study specifications; raw data acquisition; data processing and results publication.However, current PLM systems do not provide easy, flexible and user-adapted data access, especially in the context of heterogeneity expertise environments such as BMI.This paper presents VAQUERO (VisuAlization and QUERy based Ontology), a visual ontology-based data query approach, that aims at providing different kinds of users in the BMI field (common/ external, domain expert and technical users) with easy self-access to their data stored in a PLM Teamcenter system.