JOURNAL OF COMPUTER SCIENCE AND ENGINEERING, VOLUME 8, ISSUE 1, JULY 2011 40

Application of VRML and PNML for Feature Recognition of Mechanical Parts B.V.Sudheer Kumar and C.S.P.Rao Abstract---Recognition of manufacturing features forms the basis for the computer aided process planning (CAPP) and plays a key role in integration of computer aided design (CAD) and computer aided manufacturing (CAM). It is the process of converting CAD data of a part into a model of the manufacturing activities required to create the part. This paper presents feature recognition method using Virtual Reality Modeling Language (VRML) The geometric information of the part is translated into manufacturing information through VRML. A Java program in Netbeans environment is used to recognize the features, where geometric information of the part is extracted from VRML file. By using this data, an output file in the form of Petri Net Markup Language (PNML) is obtained, which is given as input to the Petri net generation software, P3. The resulting output file gives the number of machinable features present in the part in the form of places and transitions. This process has been tested on a part and successfully extracted all the features.

Index Terms— CAD, CAM, Feature Recognition, PNML, VRML

——————————
——————————

1 INTRODUCTION

F

eatures are generic shapes or other characteristics of a part with which engineers can associate knowledge useful for reasoning about the part [8]. Manufacturing features are those which are useful for manufacturing the part on the shop floor. Examples for simple manufacturing features are a hole, a pocket and a slot etc. Most of the parts can be interpreted in terms of a hole, a slot and a pocket. CAPP will use these features to generate manufacturing instructions to produce the part. For example, CAPP typically generates a drilling operation for the hole. Jung Hyun Han [4] made a survey on feature recognition and merits of several algorithms of feature recognition i.e. graph pattern matching, cell based decomposition, convex hull decomposition and Hint based reasoning etc. Mike Pratt and William C. Regli [3] gave an overview on the three major algorithmic approaches for feature recognition and mentioned several drawbacks of them. In the graph-based algorithms, the part is represented by a graph data structure, and is searched for particular patterns for features. The volumetric decomposition approach decomposes the volume of the part to be manufactured into a set of intermediate volumes and then

manipulates the volumes to produce features. The hint-based reasoning starts from a minimal indispensable portion of a feature's boundary which should be present in the part, and performs extensive geometric reasoning. Joshi and Chang [5] developed a graph named the Attribute Adjacency Graph (AAG) to represent features in which each face of the part is represented as a node, and each edge or face adjacency is represented as an arc. Sashikumar Venkataraman [6] presented a graph based frame work for feature recognition. The feature recognition step involved finding similar sub graphs in the part graph. The novelty of this framework lied in the usage of a rich set of attributes to recognize a wide range of features efficiently. W.F. Lu [7] gave an approach to recognize features from a data exchanged part model. A litany of algorithms for the identification of design and machining features are proposed. Emad S. Abouel Nasr [8] discussed a methodology for extracting manufacturing features from CAD system. The system takes a neutral file in Initial Graphics Exchange Specification (IGES) format as input and translates the information in the file to manufacturing information. The boundary (B-rep) geometrical information of the part is then analyzed by a feature recognition program that is created specifically to extract the features from the geometrical information based on a geometric reasoning approach.

————————————————

• F.A. Author is a research scholar in the Department of Mechanical Engineering, National Institute of Technology, Warangal, AP, India. PIN506021. • S.B. Author is a Professor in the Department of Mechanical Engineering, National Institute of Technology, Warangal, AP, India. PIN-506021.

The basic role of the CAD is to precisely define the geometry of the product, as it is critical to all the subsequent activities in the product life cycle. In most of the cases parts are designed in separate CAD environment, which has no direct link with

© 2011 JCSE http://sites.google.com/site/jcseuk/

JOURNAL OF COMPUTER SCIENCE AND ENGINEERING, VOLUME 8, ISSUE 1, JULY 2011 41

manufacturing. Different CAD packages use different types of database structures to store the information of the part in a CAD file. In this paper CATIA is used for design the part. However other CAD soft wares, which have an option to convert design file into Virtual Reality Modeling Language (VRML) format can also be used. Technically speaking, VRML is neither virtual reality nor a modeling language. Virtual reality typically implies an immersive 3D experience (such as a head-mounted display) and 3D input devices (such as digital gloves). VRML neither requires nor precludes immersion. Furthermore, a true modeling language would contain much richer geometric modeling primitives and mechanisms. VRML provides a bare minimum of geometric modeling features and contains numerous features far beyond the scope of a modeling language [17].Then the VRML file format will be changed to X3D format. A Java program in Netbeens environment is being used to extract the features from X3D file. To facilitate the transfer of Petri nets across various platforms, it is necessary to develop a standard transfer format. This was recognized in the initial proposal to establish an International standard for High-level Petri nets in 1995. An initiative was taken to discuss the development of a standard interchange format for Petri nets by holding a workshop as part of the Petri net conference in Aarhus, Denmark. The principles and objectives of such an interchange format were discussed and different proposals for XML-based interchange formats were presented. The Petri Net Markup Language (PNML) [21] is one of the proposals for an interchange format at the first workshop. Though not generally accepted yet, it is currently supported by a couple of tools. Moreover, it is flexible enough to integrate different types of Petri nets and is open for future extensions. This makes it a good starting point for a future standard interchange format.

multi-platform language for publishing web content in a relatively simple and straight-forward way.

2.1 The structure of VRML [18] A VRML file consist of the following main sections 2.1.1 Header The header of the file tells the VRML parser that the file is an actual VRML file of a certain version, along with the file´s encoding type. It has the following format: #VRML V2.0 utf8 [optional comment] 2.1.2 Scene graph The scene graph contains nodes which describe objects and their properties. It contains hierarchically grouped geometry to provide an audio-visual representation of objects, as well as nodes that participate in the event generation and routing mechanism.

2.1.3 Prototypes A Prototype is a node type created by the user by providing extensions to the standard set of nodes. A prototype can be seen as the struct data type in ANSI C, or as a macro.

2.1.4 Event routes Some VRML nodes generate events in response to environmental changes or user interaction. Event routes provide the creator of a scene means of catching these events and route them into other nodes that respond to these events.

2.1.5 Script nodes 2 VIRTUAL REALITY MODELING LANGUAGE VRML, sometimes pronounced vermal, is an acronym for the Virtual Reality Modeling Language. Technically speaking, VRML is neither virtual reality nor a modeling language. Virtual reality typically implies an immersive 3D experience (such as a head-mounted display) and 3D input devices (such as digital gloves). VRML neither requires nor precludes immersion. Furthermore, a true modeling language would contain much richer geometric modeling primitives and mechanisms. VRML provides a bare minimum of geometric modeling features and contains numerous features far beyond the scope of a modeling language [17]. VRML can most easily be seen as a 3D interchange format that supports common features such as hierarchical transformations, light sources, geometry, animations, visual effects, material properties and texture mapping. VRML has been designed to be an analog to the commonly used HTML, in that it is a

Script nodes are nodes used to catch and process events from other nodes. By using Scripts, more advanced functions can be achieved, rather than just the sheer passing of messages that is done by the event routing.

2.2 The VRML browser [19] The VRML files created are interpreted and presented by the browser. The browser can be either a standalone application or a plug-in for an ordinary web browser such as Internet Explorer. The browser presents the geometry of the VRML scene graph along with means of interacting, navigating and exploring the virtual world. The main components of the browser are the Parser, the Scene graph and the Presentation modules. The Parser reads the VRML file and constructs the hierarchical structure of transformations and geometry nodes comprising the scene graph from this

JOURNAL OF COMPUTER SCIENCE AND ENGINEERING, VOLUME 8, ISSUE 1, JULY 2011 42

information. The presentation module renders the scene graphically and audibly back to the user.

3 PETRI NETS Petri nets are graphical and mathematical tools provide a uniform environment for modeling, formal analysis, and design of discrete event systems. One of the major advantages of using Petri net models is that the same model is used for the analysis of behavioral properties and performance evaluation, as well as for systematic construction of discrete event simulators and controllers. Petri nets were named after Carl A. Petri who created in 1962 a net-like mathematical tool for the study of communication with automata. Their further development was facilitated by the fact that Petri nets can be used to model properties such as process synchronization, asynchronous events, concurrent operations, and conflicts or resource sharing. These properties characterize discrete-event systems whose examples include industrial automated systems, communication systems, and computer-based systems. These, and other factors discussed in this paper, make Petri nets a promising tool and technology for application to Industrial Automation [20]. Petri nets as graphical tools provide a powerful communication medium between the user, typically requirements engineer and the customer. Complex requirements specifications, instead of using ambiguous textual descriptions or mathematical notations difficult to understand by the customer, can be represented in Petri nets. This combined with the existence of computer tools allowing for interactive graphical simulation of Petri nets, puts in hands of the development engineers a powerful tool assisting in the development process of complex systems. As a mathematical tool, a Petri net model can be described by a set of linear algebraic equations, or other mathematical models reflecting the behavior of the system. This opens a possibility for the formal analysis of the model. This allows a formal check of the properties related to the behavior of the underlying system, e.g., precedence relations amongst events, concurrent operations, appropriate synchronization, freedom from deadlock, repetitive activities and mutual exclusion of shared resources, to mention some. The simulation based model validation can only produce a limited set of states of the modeled system and thus can only show presence (but not absence) of errors in the model, and its underlying requirements specification. The ability of Petri nets to verify the model formally and it is especially important for realtime safety-critical systems such as air-traffic control systems, rail-traffic control systems, nuclear reactor control systems etc. One of the most successful application areas of Petri nets has been modeling and analysis of communication protocols. The work in this area can be dated back to the early 1970s. In the past

few years, a number of approaches have been proposed which allow for the construction of Petri net models of protocols from specifications written in a relatively skill-free. Petri nets have been used extensively to model and analyze manufacturing systems. In this area, Petri nets were used to represent simple production lines with buffers, machine shops, automotive production systems, flexible manufacturing systems, automated assembly lines, resource-sharing systems, and recently just-in-time and kanban manufacturing systems.

3.1 Petri Net Markup Language (PNML) PNML has been recognized in the Petri net community for over a decade that it is useful and to be able to transfer Petri net models between tools that may exist in different countries throughout the world. This would allow Petri Net tool users in geographically distributed locations to take advantage of newly developed facilities on other tools (analysis, simulation or implementation). To facilitate the transfer of Petri nets it is useful to develop a standard transfer format. This was recognized in the initial proposal to establish an International standard for High-level Petri nets in 1995. Fifteen years ago, an initiative was taken to discuss the development of a standard interchange format for Petri nets by holding a workshop as part of the Petri net conference in Aarhus, Denmark. The principles and objectives of such an interchange format were discussed and different proposals for XML-based interchange formats were presented. The Petri Net Markup Language (PNML) [21] is one of the proposals for an interchange format at the first workshop. Though not generally accepted yet, it is currently supported by a couple of tools. Moreover, it is flexible enough to integrate different types of Petri nets and is open for future extensions. This makes it a good starting point for a future standard interchange format.

4 DEVELOPMENT OF FEATURE RECOGNITION MODULE FROM VRML FILES The part designed using the CAD forms the basis for the extraction of the features. .In this paper CATIA is used for modeling the component. The output file is saved in the VRML format. A java program is developed in the Netbeans Integrated Development Environment which is used for the extraction of the coordinates. First the VRML file is translated form the VRML file format to X3D format using the NIST’s vrml2x3d converter .The X3D file format is similar to the VRML format which represents the xml format. The IndexedLineSet and PointSet are specified by sets of coordinate points, color values and normal vectors that are specified point by point and nodes are extracted from the X3D file. Then the extraction logic is applied which results in the final file which gives all

JOURNAL OF COMPUTER SCIENCE AND ENGINEERING, VOLUME 8, ISSUE 1, JULY 2011 43

the features present (The major features which we considered in this paper are holes, therefore the final output is the radius and the centers of the circles). This output file is used to interface to the PNML code snippets. The Feature Recognition Module is well explained with example part in section -5.

5 ILLUSTRATION WITH AN EXAMPLE The integration of the diverse functions in the manufacturing has been a challenge which has been taken up and various methods have been proposed in this regard .In this paper, a new method has been adopted for the integration of CAD and CAPP. In this section the integration module is explained with example part.

5.1 CAD Module This module represents the CAD environment with functions for designing and visualization of the part. The functions of this CAD module is to allow the user to design the part in 3D and make changes if there are any changes in the design of the part. The example part i.e. the plate with holes which has been modeled in CATIA is shown in Fig 1. Any other CAD package like Pro-E, IDEAS can also be used. The output file is saved in VRML file format and this output file is given as input to the next module.

Fig. no. 2 The Feature Recognition Module in the Netbeans environment. The coordinates are grouped so that they represent the features altogether. The function of the feature extraction module is to extract the coordinates which represent the features present in the part from the VRML file generated form the CAD module. The VRML file is converted in X3D format which is an XML based format which represents the new generation neutral file format for data translation. The vrml2x3d converter developed by NIST (USA) is used for this purpose.

Fig. no. 1 CAD Module

5.2 Feature extraction from the VRML file This module represents the Feature extraction from the VRML file of the part and the PNML file format generation (Fig.2). The module is developed in Netbeans environment and the codes for extraction of feature representing coordinates are developed in Java.

Fig. no. 3. The VRML browser (OCTAGA PLAYER) The X3D file can be viewed on OCTAGAPLAYER shown in Fig.3, which has been used for other functions like distribution of data between the other departments in the manufacturing facility and also between the facilities outside the firm. The logic developed in Java is used to recognize the features. For the above example part the out put from the java program is shown in table.1 for concentric holes having different diameters (55 & 50 units) with center (-160, 110, Z). The first hole (R = 28.5) having upper

JOURNAL OF COMPUTER SCIENCE AND ENGINEERING, VOLUME 8, ISSUE 1, JULY 2011 44

circle at Z = 0 and lower circle at Z = 8 and these two circles passes through the six Cartesian Points (CP) having same X and Y coordinates is shown below. Similarly the second hole (R = 22.5) having upper circle at Z = 8 and lower circle at Z = 30. Similarly all hole features are identified in the example part and these features are connected to the respective PNML code snippets which represent the final output file in PNML syntax .This output is given as input for the next module.

[2]

[3]

[4]

[5]

[6] [7]

[8]

[9]

[10]

[11]

[12]

[13]

Fig. no.7. The PNML output file format [14]

6 CONCLUSIONS This paper proposes a novel approach towards the integration of CAD and CAPP. The geometrical data obtained from a CAD model is saved in VRML file format. Then the VRML file format will be changed to X3D format. A Java program in Netbeens environment is being used to extract the features from X3d file. The features obtained from the java program will be converted to PNML file format By using this approach manufacturing feature recognition of components can be done using VRML file format, and the CAD model is seamlessly translated into a Petri model which represents the features of the part.

[15]

[16]

[17] [18] [19]

7 REFERENCES

[20]

[1]

[21]

Nafis Ahmad, and A.F.M. Anwarul Haque., “Manufacturing feature recognition of parts Using DXF file,” 4th International Conference on Mechanical Engineering, pp. VI 111-115,

December 2001, Dhaka, Bangladesh. (Conference Proceedings) T. Dereli, and H. I. Filiz, “Optimization of Process Planning functions by Genetic algorithm,” Vol. 36, pp281-308, 1999, (Computers and Industrial Engineering) Mike Pratt and William C. Regli, “Manufacturing Feature Recognition from Solid Models: A Status Report,” vol. 16, no. 6, pp 782-796, 2000 (IEEE transactions on robotics and automation) Jung Hyun Han, “Survey on Feature Research,” Institute for Robotics and Intelligent systems, USC, USA, 1996. (Technical report IRIS-96-346) S. Joshi, and T. C. Chang, “Graph Based Heuristics for Recognition of Machined Features from 3D Solid Model,” Vol. 20, pp. 58-66, 1988. (Computer-Aided Design) Sashikumar Venkataraman, “A Graph based Frame work for Feature recognition,” ACM 2001 1-58113-366-9/01/06, 2001. W.F. Lu, “An approach to identify design and manufacturing features from data exchanged part model,” Computer-Aided Design 35 (2003) 979–993, 2003. Emad S. Abouel Nasr, “A new methodology for extracting manufacturing features from CAD system,” Computers & Industrial Engineering 51, 389–415, 2006. Murata Tadao, “Petri Nets: Properties, Analysis and applications.” 1989. (IEEE Transactions on Electrical Engineering and Computer Science). K. Srihari, and C. R. Emerson, “Petri nets in dynamic process planning,” 19:447–451, 1990. (Computers & Industrial Engineering) J. A. Cecil, C. R. Emerson, K. Srihari, “A review of Petri net applications in process planning,” 7:168–177, 1992. (International Journal of Advanced Manufacturing Technology) P. Xirouchakis and D. Kiritsis, “A generic Petri net model for disassembly planning processes,” presented at TMCE’96, Budapest, 1996 D. Kiritsis and M. Porchet, “A generic Petri net model for dynamic process planning and sequence optimization.” Adv. Engineering Software, 25(1):61–71, 1996. D. Kiritsis, K. P. Neuendorf, and P. Xirouchakis, ”Petri net techniques for process planning cost estimation,” Advances in Engineering Software, 30, 375–387, 1999. J. M. Usher, and R. O. Bowden, “The application of genetic algorithms to operation sequencing for use in computer-aided process planning,” Computers & Industrial Engineering, ISSN 0360-8352, Vol. 30, No.4, pp. 999-1013, 1996. H.B.Marri, A. Gunasekaran, and R. J. Grieve, “Computer aided process planning: A state of art survey,” International journal of Manufacturing Technology, 14: 261-268, 1998. Ramesh Naidu Ande, Integrated VRML, JAVA, XML and HTML in a web based tool, SCSC, 2004. Don Brutzman, The Virtual Reality Modeling Language and Java, Communications of the ACM, vol. 41 no. 6, pp. 57-64, 1998. Kian-Huat Tan, Tze Leong Yew, and Kurt Gramoll, “Understanding Machine Operations and Manufacturing using VRML,” Am. Soc. of Engineering Educ, (ASEE) 1999 Conf., June 20 - 23, Charlotte, NC, 1999. Anil Sawhney, Andre Mund, and Jennifer Marble, “Simulation Of The Structural Steel Erection Process,” Proceedings of the 31st Winter Simulation Conference, 1999. WD Version 0.9.0, Software and Systems Engineering – Highlevel Petri Nets Part 2: Transfer Format, International Standard ISO/IEC 15909-2 June 23, 2005.

JOURNAL OF COMPUTER SCIENCE AND ENGINEERING, VOLUME 8, ISSUE 1, JULY 2011 45

Dr. C. S. P. Rao Obtained his B.Tech degree from JNTU Kakinada in 1985, M.Tech from RECW in 1988, Ph.D from RECW in 1997, PG Diploma in Management in 2003 and M.B.A. in 2005. After working in DRDO as Scientist ‘B’, joined in REC Warangal. Presently working as Professor in Mechanical Department, National Institute of Technology, Warangal (formerly known as RECW). Published 38 papers in international journals, and 10 papers in national journals. Presented 166 papers in international and national conferences. In 2008, got the Andhra Pradesh Scientist Award and also Engineer of the Year Award. He is reviewer for 5 international journals.

B.V.Sudheer Kumar Obtained B.Tech degree in 1992 from Bapatla Engineering College, M.Tech from NITW in2003, and pursuing Ph.D from NITW. After working 6 years in industry, since 1998 continuing in the teaching field and worked in various colleges. The list in sequence starts with 3 years in Vazir Sultan college of Egineering, Khammam, 3.5 years in Gudlavalleru Engineering College, Gudlavalleru, 1 year in Supraja Institute of Technology & Science, Warangal, presently with NIMRA Group of colleges from last 1.5 years, heading the Department of Mechanical Engineering, Nimra College of Engineering & Technology, Vijayawada. Presented papers in 1 international conference FAIM ’09 in UK and 3 national conferences in India.

Table 1. Coordinate data from feature recognition module Radius Z

CP1 X

CP2 Y

X

CP3 Y

X

CP4 Y

X

CP5 Y

X

CP6 Y

X

Y

28.5

0

-185 96 -181 91 -177 87 -134 123

-139 129 -143 133

28.5

8

-185 96 -181 91 -177 87 -134 123

-139 129 -143 133

22.5

8

-180 99 -176 94 -172 91 -140 121

-144 126 -148 129

22.5

30 -180 99 -176 94 -172 91 -140 121

-144 126 -148 129