Databases of 3D CAD (computer aided design) models are often large and lacking in meaningful organisation. Effective tools for automatically searching for, categorising and comparing CAD models, ...therefore, have many potential applications in improving efficiency within design processes. This paper presents a novel asymmetric autoencoder model, consisting of a recursive encoder network and fully-connected decoder network, for the reproduction of CAD models through prediction of the parameters necessary to generate a 3D part design. Inputs to the autoencoder are STEP (standard for the exchange of product data) files, an ISO standard CAD model format, compatible with all major CAD software. A complete 3D model can be accurately reproduced using a STEP file, meaning that all geometric information can be used to contribute to the final encoded vector, with no loss of small detail.
In a CAD model of overall size 10 × 10 × 10 units, for 90% of models, the class of an added feature is estimated with maximum error of 0.6 units, feature size with maximum error of 0.4 units and coordinate values representing position with maximum error of 0.3 units. These results demonstrate the successful encoding of complex geometric information, beyond merely the shape of the 3D object, with potential application in the design of search engine functionality.
Currently process engineers are using documents or authoring tools to bring the assembly instructions to the work floor. This is a time-consuming task, as instructions need to be created for each ...assembly operation. Furthermore, the engineer needs to be familiar with the assembly sequence. To assist the engineer, a tool is developed that i) uses a heuristic based on visibility, part similarity and proximity to semi-automatically determine the assembly sequence from a CAD model and ii) according to the computed sequence generates digital work instructions including visualizations and animations extracted from the CAD model. In essence, the assembly sequence generation works reversely: it determines the order in which components can be removed from the assembly, by evaluating whether the visibility of a component is obstructed by the remaining assembly. The reversed order is then returned as assembly sequence. During this process the engineer can modify the proposed sequence, add annotations and alter the visualizations of the proposed instructions, i.e., images or 3D-animations. We illustrate that the developed tool effectively supports process engineers and speeds up the creation of digital work instructions by some industrial validation cases, e.g., the assembly of a weaving machine.
3D printing via reversible addition‐fragmentation chain transfer (RAFT) polymerization has been recently developed to expand the scope of 3D printing technologies. A potentially high‐impact but ...relatively unexplored opportunity that can be provided by RAFT‐mediated 3D printing is a pathway toward personalized medicine through manufacturing bespoke drug delivery systems (DDSs). Herein, 3D printing of drug‐eluting systems with precise geometry, size, drug dosage, and release duration/profiles is reported. This is achieved through engineering a range of 3D models with precise interconnected channel‐pore structure and geometric proportions in architectural patterns. Notably, the application of the RAFT process is crucial in manufacturing materials with highly resolved macroscale features by confining curing to exposure precincts. This approach also allows spatiotemporal control of the drug loading and compositions within different layers of the scaffolds. The ratio between the polyethylene glycol units and the acrylate units in the crosslinkers is found to be a critical factor, with a higher ratio increasing swelling capacity, and thus enhancing the drug release profile, from the drug‐eluting systems. This proof‐of‐concept research demonstrates that RAFT‐mediated 3D printing enables the production of personalized drug delivery materials, providing a pathway to replace the “one‐size‐fits‐all” approach in traditional health care.
Herein, 3D printing of a customized drug‐eluting system via reversible addition‐fragmentation chain transfer (RAFT) polymerization is presented. This research demonstrates that RAFT‐mediated 3D printing enables the production of drug delivery materials with controlled architecture, size, drug dosage, and release profiles. This system provides a pathway toward personalized medicine by producing bespoke drug delivery systems.
Increasingly complex 3D CAD models are essential during different life-cycle stages of modern engineering projects. Even though these models contain several repeated geometries, instancing ...information is often not available, resulting in increased requirements for storage, transmission, and rendering. Previous research have successfully applied shape matching techniques to identify repeated geometries and thus reduce memory requirements and improve rendering performance. However, these approaches require consistent vertex topology, prior knowledge about the scene, and/or the laborious creation of labeled datasets. In this paper, we present an unsupervised deep-learning method that overcomes these limitations and is capable of identifying repeated geometries and computing their instancing transformations. The method also guarantees a maximum visual error and preserves intrinsic characteristics of surfaces. Results on real-world 3D CAD models demonstrate the effectiveness of our approach: the datasets are reduced by up to 83.93% in size. Our approach achieves better results than previous work that does not rely on supervised learning. The proposed method is applicable to any kind of 3D scene and geometry.
Display omitted
•Unsupervised deep learning method for 3D shape registration.•Does not require any previous knowledge of the 3D geometries.•Does not require a labeled dataset for any supervised training.•Guarantees an upper bound on any visual errors.•Generalizes for any 3D scene and geometry.
Currently in industry, design and communication of a product assembly is through the use of computer-aided design (CAD) systems. However, there are no commercial systems that can automatically ...generate feasible assembly sequence plans. There is past and current academic research in methods to provide automatic assembly sequence planning. Assembly sequence planning using a commercial system often relies on an expert assembly sequence planner, and it is predominantly done manually. This requires a great amount of time and expert knowledge; assembly sequence plans generated may not even be the most efficient. The ability to automatically generate assembly sequence plans will lead to the reduction of planning time, less reliance on the amount of knowledge required, and better plans at earlier stages of the design process. CAD models are based on feature constraints to create and define an assembly. The challenges to automatically generate assembly sequences using CAD models lie in intelligent reasoning and analysis of the modelled assembly data. Based on past research findings, there is a reason to believe that assembly constraints used in CAD assembly models can provide essential information related to the assembly process. This paper presents a system that can analyse and utilize assembly data available from a CAD model to generate assembly sequences. The system also considers a user input as a type of assembly constraint. The system is capable of producing a set of ranked feasible assembly sequence plans for an operator to evaluate. A matrix approach has been adopted to process the information retained from a CAD model. Interference and stability studies are carried out during the creation of assembly sequence plans. The outputs are ranked based on the ease of assembly and the stability of the generated assembly sequence plans. Case studies are used to evaluate the system and the feasibility of the output. A case study using a two stroke engine is presented, which demonstrates how the system generates assembly sequence plans.
•Automatic collection of constraint data from CAD models for assembly planning.•Algorithm for generating assembly sequences based on the CAD constraints.•Generated assembly sequences are validated via interference and stability analysis.•System able to be used with any modern CAD system, with demonstration in CREO.
This paper poses object category detection in images as a type of 2D-to-3D alignment problem, utilizing the large quantities of 3D CAD models that have been made publicly available online. Using the ..."chair" class as a running example, we propose an exemplar-based 3D category representation, which can explicitly model chairs of different styles as well as the large variation in viewpoint. We develop an approach to establish part-based correspondences between 3D CAD models and real photographs. This is achieved by (i) representing each 3D model using a set of view-dependent mid-level visual elements learned from synthesized views in a discriminative fashion, (ii) carefully calibrating the individual element detectors on a common dataset of negative images, and (iii) matching visual elements to the test image allowing for small mutual deformations but preserving the viewpoint and style constraints. We demonstrate the ability of our system to align 3D models with 2D objects in the challenging PASCAL VOC images, which depict a wide variety of chairs in complex scenes.
Search and retrieval remains a major research topic in several domains, including computer graphics, computer vision, engineering design, etc. A search engine requires primarily an input search query ...and a database of items to search from. In engineering, which is the primary context of this paper, the database consists of 3D CAD models, such as washers, pistons, connecting rods, etc. A query from a user is typically in the form of a sketch, which attempts to capture the details of a 3D model. However, sketches have certain typical defects such as gaps, over-drawn portions (multi-strokes), etc. Since the retrieved results are only as good as the input query, sketches need cleaning-up and enhancement for better retrieval results.
In this paper, a deep learning approach is proposed to improve or clean the query sketches. Initially, sketches from various categories are analysed in order to understand the many possible defects that may occur. A dataset of cleaned-up or enhanced query sketches is then created based on an understanding of these defects. Consequently, an end-to-end training of a deep neural network is carried out in order to provide a mapping between the defective and the clean sketches. This network takes the defective query sketch as the input and generates a clean or an enhanced query sketch. Qualitative and quantitative comparisons of the proposed approach with other state-of-the-art techniques show that the proposed approach is effective. The results of the search engine are reported using both the defective and enhanced query sketches, and it is shown that using the enhanced query sketches from the developed approach yields improved search results.
Display omitted
•The first learning-based strategy to clean rough query sketches of 3D CAD models•Introduces SketchCleanNet — an end-to-end image translation scheme•SketchCleanNet aims to understand the mapping between rough sketches and clean query images•A novel scheme to calculate the loss is introduced•Dataset Contribution: The resulting enhanced query sketch dataset is made available publicly.•This paper will significantly contribute to the research community and give researchers opportunities to develop new algorithms for search and retrieval of 3D mechanical components.
As Virtual Reality(VR), Augmented Reality(AR), Mixed Reality(MR) technology becomes more accessible, it is important to explore VR/AR/MR technologies that can be used for remote collaboration on ...physical tasks. Previous research has shown that gesture-based interaction is intuitive and expressive for remote collaboration, and using 3D CAD models can provide clear instructions for assembly tasks. In this paper, therefore, we describe a new MR remote collaboration system which combines the use of gesture and CAD models in a complementary manner. The prototype system enables a remote expert in VR to provide instructions based on 3D gesture and CAD models (3DGAM) for a local worker who uses AR to see these instructions. Using this interface, we conducted a formal user study to explore the effect of sharing 3D gesture and CAD models in an assembly training task. We found that the combination of 3D gesture and CAD models can improve remote collaboration on an assembly task with respect to the performance time and user experience. Finally, we provide some conclusions and directions for future research.
Efficient and effective construction progress tracking is critical to construction management. Current manual tracking methods are time consuming and/or error prone. Three dimensional (3D) laser ...scanners are being investigated in the construction industry and have shown potential for supporting progress tracking. However, their full potential has not yet been achieved. The reason may be that commercial software packages are still too complicated for processing scanned data. Methods have however been developed for the automated recognition of project 3D CAD model objects in site laser scans. A novel system is thus described herein that combines 3D object recognition technology with schedule information into a combined 4D object oriented progress tracking system. This system is tested on a comprehensive field database acquired during the construction of the Engineering V Building at the University of Waterloo. It demonstrates a degree of accuracy for automated progress tracking that meets or exceeds typical manual performance.
► 3D Laser scans are fused with 4D models to automate construction progress control. ► The system is tested with the data collected from a concrete building construction. ► The results indicate the importance of planning for scanning. ► Using updated schedules gives better progress estimation results.