Geometrical reasoning based on attributed graph grammar for prismatic parts

Modern manufacturing systems rely on a smooth and quick transformation of the design specification into manufacturing instructions. The shift towards feature-based design and manufacturing supports this effort. One of the weak links in this chain of activities is the conversion of the design features into the manufacturing features. This article presents a method based on graph grammar parsing that converts design geometrical features into manufacturing geometrical features. These features can then be used in process planning activities. The method presented extends the state of the art of syntactic graph-based features recognition by supporting embedding transformations, attribute transfer algorithms, decomposition of overlapping features and removal of empty volumes. In addition, the analysis provided by the graph parsing algorithms implicitly treats the features as solid models without the need to perform complicated geometrical'analysis. The system was implemented for prismatic components such that they can be produced on a triple axis vertical milling machine, and currently contains 20 rules.     

Constraint-based design using an operational approach

Constraint-based design, which explicitly represents and operates upon constraints, has been recognized as a promising tool for achieving intelligent support of design, particularly the design of mechanical parts or assemblies. It is essential for a constraint-based system to realize the constraint-solving capability. This paper presents an operational approach to constraint solving using incremental feature operations. The approach is based on an operational interpretation of constraints, i.e. the constraint satisfaction is carried out in terms of operations incrementally. A grammatic formalism is used for operational modeling of constraints. Each graph production within a graph grammar corresponds to an operation or a sequence of operations designated for constraint satisfaction that is related to a rule or a procedure. Therefore, a constraint satisfaction process can be represented by a graph grammar parsing process. The operation sequence is planned by graph grammar parsing and invocation of the related rules or procedures. Constraints are then evaluated by invoking the sequenced operations. Features are introduced as higher-level abstractions into the geometric constraints network. This enables reasoning about design validation from topological and manufacturing views.     

Recognition of machining features - a hybrid approach

This paper describes a hybrid system which endeavours to recognize machining features automatically from a boundary representation (b-rep)-based solid modeller. The graph-based approach and the volume approach are adopted in consecutive stages in a prototype feature recognition system to combine the positive aspects of both strategies. The graph-based approach is based on feature edge sequence (FES) graph, a new graph structure introduced in this system. The FES graph approach is used to extract primitive features from the three-dimensional solid model; and the volume decomposition approach is incorporated to generate multiple interpretations of the feature sets. In addition, a neural network (NN)-based technique is used to tackle the problem of nonorthogonal and arbitrary features. Using the hybrid system, a workpiece designed in b-rep solid modeller will be interpreted and represented by a set of primitive features attached with significant manufacturing parameters, including multiple interpretations, tool directions and machining sequences, etc. The overall hybrid system is able to transform a pure geometric model into a machining feature-based model which is directly applicable for downstream manufacturing applications.     

A hybrid method for recognizing interacting machining features

Recognizing interacting features from a design part is a major challenge in the feature recognition problem. It is difficult to solve this problem using a single reasoning approach or artificial intelligence technique. A hybrid method, which is based on feature hints, graph theory and an artificial neural network--ART 2 net--has been proposed to recognize interacting machining features. Through enhancing the concepts of feature hints and graph representation schemes, which were presented in previous work to facilitate the extraction process of interacting features and reduce the searching space of recognition algorithms, a novel set of representations and methodologies to define generic feature hints (F-Loops), the interacting relationships between F-Loops and graph manipulations for F-Loops are developed to deduce potential features with various interacting relationships in a unified way. The obtained potential features are represented as F-Loop Graphs (FLGs), and these FLGs are input into an ART 2 neural network to be classified into different types of features eventually. The advantages of employing the ART 2 network are highlighted through comparing the computational results with another type of neural network, which is commonly utilized in the feature recognition domain. Case studies with complex interacting features show that the developed hybrid method can achieve optimal efficiency by benefiting from the diverse capabilities of the three techniques in the different phases of the recognition approach.     

Recognition of interacting rotational and prismatic machining features from 3-D mill-turn parts

Current feature recognition methods generally recognize and classify machining features into two classes: rotational features and prismatic features. Based on the different characteristics of geometric shapes and machining methods, rotational features and prismatic features are recognized using different methods. Typically, rotational features are recognized using two-dimensional (2-D) edge and profile patterns. Prismatic features are recognized using 3-D geometric characteristics, for example, patterns in solid models such as 3-D face adjacency relationships. However, the current existing feature recognition methods cannot be applied directly to a class of so-called mill-turn parts where interactions between rotational and prismatic features exist. This paper extends the feature recognition domain to include this class of parts with interacting rotational and prismatic features. A new approach, called the machining volume generation method, is developed. The feature volumes are generated by sweeping boundary faces along a direction determined by the type of machining operation. Different types of machining features can be recognized by generating different forms of machining volumes using various machining operations. The generated machining volumes are then classified using face adjacency relationships of the bounding faces. The algorithms are executed in four steps, classification of faces, determining machining zones, generation of rotational machining volumes and prismatic machining volumes, and classification of features. The algorithms are implemented using the 3-D boundary representation data modelled on the ACIS solid modeller. Example parts are used to demonstrate the developed feature recognition method.     

A system for automatic extraction of 3D part features using syntactic pattern recognition techniques

This paper reports a system which employs string based syntactic pattern recognition techniques for feature extraction. A B-rep based solid model of the component is used for processing the face boundaries (loops) to derive boundary strings which are syntactically analysed to detect feature signatures. The detected signatures are geometrically verified, and the positional and dimensional parameters of the features are calculated. The system is capable of detecting features with complex cross sections; compound features which include nested and intersecting features; and various types of holes and pockets. These capabilities of the system make it an effective interface between a geometric solid modeller and various application programs.     

A framework for CAD-based geometric reasoning for robot assembly language†

For automatic robot programming, world modelling of the robot's environment is one of the most important phases of the task planning. World modelling requires that the robot know the environment in which it operates, including the spatial configuration of all objects in the task environment. In robotic assembly planning at the task level, representation of these objects requires symbolic feature and shape identification of the objects to be assembled by the robot. In this paper, we present a framework for reasoning about objects based on their shapes and features and the representation of such objects for robotic assembly planning when the modelling is done on a CAD system. We show the importance of AI languages in the communication of constructive solid geometry (CSG) based information from modellers. Finally, we present the schematic for a formalism, based on Prolog, for expressing object properties and assembly situations.     

基于改进双流卷积递归神经网络的RGB-D物体识别方法

为了提高基于图像的物体识别准确率,提出一种改进双流卷积递归神经网络的RGB-D物体识别算法(Re-CRNN).将RGB图像与深度光学信息结合,基于残差学习对双流卷积神经网络(CNN)进行改进:增加顶层特征融合单元,在RGB图像和深度图像中学习联合特征,将提取的RGB和深度图像的高层次特征进行跨通道信息融合,继而使用So...     

A shape grammar for non-manifold modeling

Shape grammars offer a notationally rich representation for dealing with shapes defined over an algebra of points, lines, planes, and solids. Operations in the algebra are bound in subshape recognition and replacement, making them ideal candidates for design as formal solid modeling representations, and for manufacturing as shape-based feature recognizers. As well, given the topological hierarchy of the algebra of shapes, non-manifold modeling is clearly a fundamental part of shape grammars. Thus, one can work at the level of wireframe models, boundary models, and solid models. These characteristics make the shape grammars eminently suitable as a formal representation for both manifold and non-manifold representations of discrete shape.     

Feature vector: a graph-based feature recognition methodology

Recognition of machining features is a vital link for the effective integration of various modules of computer integrated manufacturing systems (CIMS). Graph-based recognition is the most researched method due to the sound mathematical background of graph theory and a graph's structural similarity with B-Rep computer-aided design modellers’ database. The method, however, is criticized for its high computational requirement of graph matching, its difficulty in building a feature template library, its ability to handle only polyhedral parts and its inability to handle interacting features. The paper reports a new edge classification scheme to extend the graph-based algorithms to handle test parts with curved faces. A unique method of representing a feature, called a feature vector, is developed. The feature vector generation heuristic results in a recognition system with polynomial time complexity for any arbitrary attributed adjacency graph. The feature vector can be generated automatically from B-Rep modellers. This helps in building incrementally a feature library as per the requirements of the specific domain. The proposed system is implemented in VC++ using an ACIS® 3D solid modelling toolkit.     

A methodology for recognizing features in two-dimensional cylindrical part designs

The interfacing of computer-aided design (CAD) to computer-aided manufacturing (CAM) is a vital step in automated manufacturing. An essential operation is the recognition of features from the part design. This paper presents a methodology for the recognition of features from two-dimensional rotational objects. First, this work defines the term ‘feature’ as a set of connected lines in the profile of the object, which satisfy certain geometric properties. Then, the task of feature recognition is decomposed into a set of distinct functions. These recognize, classify, decompose and reconstruct, and identify face sets which satisfy the definition of features. A prototype is developed which implements these functions. The important characteristics of this methodology are: (1) all cylindrical features are recognized, and most of them identified according to the input formats of a desired CAPP system; and (2) the system is modular and flexible and its functions can be easily modified.     

一种基于视图和支持向量机的三维物体识别方法

为提高三维物体识别系统性能并减少计算复杂性,本文提出了一种基于视图的方法.首先从三维物体的二维视图中提取颜色矩、纹理特征和仿射不变矩.颜色矩对于物体的大小和姿态不敏感且性能稳健.纹理特征可区别形状相似但外观不同的物体.仿射不变矩在物体发生仿射形变下具有不变性.本文将上述各种特征组合为23个分量的特征向量,送入支持向量机进行训练并识别.基于两种公开的三维物体数据库COIL-100和ALOI测试了本文方法性能.当每物体训练视角为36个(视角间隔10°)时,在两个数据库上的实验都达到了100%的识别率.进一步减少训练视角数量也达到较满意的识别性能,优于文献中的方法.     

Recognition and interpretation of interacting and non-interacting features using spatial decomposition and Hamiltonian path search

Since it is a complex task to formalize the feature recognition problem explicitly, a large variety of systems has been developed. One of the problems these systems have to overcome is the recognition and interpretation of interacting features. A fair success has been achieved in surface based methods to recognize certain classes of interacting features. However the problem remains for general cases of interacting features. Recently much effort has been focused on the volumetric approach. We present here the current state of a volumetric feature recognition method. The system considers interacting features in prismoidal parts and it operates in two stages: (1) recognition of regions of interest: a spatial decomposition of the space bounded by a predefined circumscribing volume is performed. A ‘cell evaluated and directed adjacency graph’ is then established. This graph is traversed to identify cavity volumes. (2) interpretation: cavity volumes made up of more than one cell can be produced by different machining operations. A graph-based decomposition method and Hamiltonian path search are combined to generate interpretations which correspond to optimal machining. The system CEDAG developed in this work uses a cell-face directed graph and contrasts the face-edge and edge-vertex graphs encountered in most conventional graph-based recognition methods.     

融合2D-PCA及稀疏表示的掌纹识别方法

提出一种基于稀疏表示的掌纹识别方法,该方法借鉴二维主成分分析(PCA)良好的数据压缩属性和较快的特征提取速度,生成掌纹特征图像.二维PCA不仅克服了一维PCA数据维数过大不易计算的缺点,而且保留了原始图像的数据结构,提取的特征能更好的代表原始图像.为了便于稀疏表达,对提取的掌纹特征图像利用一维主成分分析进行二次特征提取,得到训练样本.虽然此处使用了一维PCA,但是由于这是二次特征提取,提取的特征还是保留了原始图像的数据结构,相比单纯的一维PCA,提高了识别率.利用训练样本构造出冗余字典,并采用稀疏表示理论将测试样本表示为字典原子的线性组合,然后根据表示系数的稀疏性与稀疏集中度实现分类识别.由于该方法利用了表达系数的稀疏性,因此减小了算法的时间和空间复杂度.实验表明,针对香港理工大学的MSpalmprints Database,本文方法的识别率较传统方法有明显提高.     

Shape detection with vision: implementing shape grammars in conceptual design

Despite more than 30 years of research, shape grammar implementations have limited functionality. This is largely due to the difficult problem of subshape detection. Previous research has addressed this problem analytically and has proposed solutions that directly compare geometric representations of shapes. Typically, such work has concentrated on shapes composed of limited geometry, for example straight lines or parametric curves, and as a result, their application has been restricted. The problem of general subshape detection has not been resolved. In this paper, an alternative approach is proposed, in which subshape detection is viewed as a problem of object recognition, a sub-domain of computer vision. In particular, a general method of subshape detection is introduced based on the Hausdorff distance. The approach is not limited in terms of geometry, and any shapes that can be represented in an image can be compared according to the subshape relation. Based on this approach, a prototype shape grammar system has been built in which the geometry of two-dimensional shapes is not restricted. The system automates the discovery of subshapes in a shape, enabling the implementation of shape rules in a shape grammar. Application of the system is illustrated via consideration of shape exploration in conceptual design. The manipulations of sketched design concepts are formalised by shape rules that reflect the types of shape transformations employed by designers when sketching.     

Topological and parametric optimization of gear trains

A method for automating the design of gear trains comprised of simple, compound, bevel and worm is described. The search process combines topological changes, discrete variable choices and continuous variable optimization. By combing best-first search, implicit enumeration, automated optimization invocation and gradient-based optimization, a near guarantee of the optimal solution can be made. While the combination of methods is specific to gear trains, there are aspects of the work that make it amenable to other engineering design problems. In addition, the topological and discrete modifications to the candidate solutions are specific to gear trains, but the graph grammar methodology that is adopted has been tailored to other problems. This article presents details on the rules that generate feasible gear trains, the evaluation routines used in determining the objective functions and constraints, and the interaction among the three search methods. Resulting gear trains are presented for a variety of gear problems.     

基于分层形状文法的蜡染花朵纹创新设计研究

目的为促进蜡染这一民族传统艺术与现代审美的有机结合,提升蜡染纹样的文化影响力和传播力,提出一种基于分层形状文法的计算机辅助传统纹样设计方法,保留传统蜡染纹样的视觉特性同时满足现代审美,使传统蜡染焕发出新的生机。方法首先分析贵州传统蜡染纹样中的花朵纹造型特征,提取造型元素并分层,然后运用分层形状文法进行纹样推理演变;依据提出的创新方法生成大量不同形状,目标形状与结果形状通过规则连接得到整个形状衍生树模型。最后设计师根据自身审美和经验对形状衍生树节点进行人机交互选择,修剪衍生树,排除不符合条件的节点,完成设计方案的决策。结论改进形状文法使其适用于蜡染造型设计,建立纹样衍生树模型,并利用传统蜡染花朵纹的创新设计,验证该方法的可行性和有效性。     

Feature patterns in recognizing non-interacting and interacting primitive,circular and slanting features using a neural network

Many varied techniques have long been suggested for the recognition of features from solid modellers, and the systems which have incorporated these techniques have achieved a moderate success. However the problem of recognition of the wide variety of features, e.g. interacting and non-interacting primitive, circular and slanting features, that any real life component may have, requires complex systems which are inflexible and hence limited in their use. Here, we present a simple and flexible system in which the features are defined as patterns of edges and vertices to deal with all the above types of features. The system starts by searching a B-rep solid model, using a cross-sectional layer method, for volumes which can be considered to represent features. Once the volumes are detected, their edges and vertices are processed and arranged into feature patterns which are used as input for a neural network to recognize the features. Simple conventions used in this work enable the creation of feature patterns for primitive, circular and slanting features. Learning, generalizing and tolerating incomplete data are some of the neural network's attributes exploited in this work to deal with interacting and non-interacting features.     

Unsupervised Log Anomaly Detection Method Based on Multi-Feature

Log anomaly detection is an important paradigm for system troubleshooting. Existing log anomaly detection based on Long Short-Term Memory (LSTM) networks is time-consuming to handle long sequences. Transformer model is introduced to promote efficiency. However, most existing Transformer-based log anomaly detection methods convert unstructured log messages into structured templates by log parsing, which introduces parsing errors. They only extract simple semantic feature, which ignores other features, and are generally supervised, relying on the amount of labeled data. To overcome the limitations of existing methods, this paper proposes a novel unsupervised log anomaly detection method based on multi-feature (UMFLog). UMFLog includes two sub-models to consider two kinds of features: semantic feature and statistical feature, respectively. UMFLog applies the log original content with detailed parameters instead of templates or template IDs to avoid log parsing errors. In the first sub-model, UMFLog uses Bidirectional Encoder Representations from Transformers (BERT) instead of random initialization to extract effective semantic feature, and an unsupervised hypersphere-based Transformer model to learn compact log sequence representations and obtain anomaly candidates. In the second sub-model, UMFLog exploits a statistical feature-based Variational Autoencoder (VAE) about word occurrence times to identify the final anomaly from anomaly candidates. Extensive experiments and evaluations are conducted on three real public log datasets. The results show that UMFLog significantly improves F1-scores compared to the state-of-the-art (SOTA) methods because of the multi-feature.