Data processing: measuring – calibrating – or testing – Measurement system – Dimensional determination
Reexamination Certificate
2001-04-20
2003-01-28
Hilten, John S. (Department: 2863)
Data processing: measuring, calibrating, or testing
Measurement system
Dimensional determination
C345S418000, C345S419000, C345S420000, C345S441000
Reexamination Certificate
active
06512995
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates to a shape analysis system for analyzing shape data for an electronic product, and for comparing the shapes of multiple products or for searching for a product having a similar shape.
BACKGROUND ART
At the present, the general rule in manufacturing circles is for a variety of information entries for a product to be electronically stored and managed in a database. In addition, such a database is also employed for a product for the electronic management of detailed shape information, that is, shape data, for a solid model or a surface model (hereinafter collectively referred to as a solid/surface model), and the configuration of the product's parts.
In the manufacturing business, PDM (Product Data Management) or VPDM (Visual Product Data Management) is employed for the management of the solid/surface models of products. However, before such product data can be registered in a system, management of a solid/surface model is the personal responsibility of a designer working in a design department. Therefore, the updating of the data that accompanies a design change must be performed manually by the designer.
Generally, a plurality of departments, such as a design department and a manufacturing department, are involved in the development of a product. Therefore, before product data for a specific product are stored in a PDM or VPDM database, the data may scattered among the databases used by the respective departments. Especially when a different CAD system is employed for the design of a machine to fill a predetermined job order for another, nonaffiliated company, the product data tends to be distributed among several databases. In this case, the updating of data in a specific department must be synchronized with the updating of like data stored in the databases of other departments. Conventionally, this updating operation has also been a manual activity.
Further, in the manufacturing business the effective re-use of old design data is of prime importance when a design job is undertaken.
Therefore, a management method using parts numbers is appropriate for standard parts, a set of which is standardized to a degree and can be systematically sorted using part numbers. Since a systematic management process can be implemented using parts numbers, appropriate parts can be easily selected when old design data are to be used for a design operation.
However, the systematic management of part numbers is difficult when there are nonstandard parts that have been derived from standard parts, as needed, or that have been newly designed. Therefore, even when a database using a solid model or a surface model is employed to manage parts, such as nonstandard parts that are not systematically managed, the updating of databases and data adjustments tend to be delayed. Therefore, when old design data for nonstandard parts are employed in a design process, an appropriate part can not easily be located. And as a result, a vicious cycle has arisen, in which each interested designer strives to fashion a new part to reduce the labor required to cull a desired part from among those intended for old products, and accordingly, there is an increase in the types of nonstandard parts that are available.
PROBLEMS TO BE SOLVED BY THE INVENTION
As is described above, for electronic product management in the manufacturing business, conventionally, an updating operation that includes the storage in a database of detailed changes applied during a design process, and the synchronized updating of like data in multiple databases are manually performed, so that an artificial erroneous operation may be result.
In the manufacturing business, especially when cooperative work is engaged in by an upstream department and a downstream department, if an erroneous operation for the preparation or the changing of a solid/surface model used in common occurs in the upstream department, enormous damage may occur. As an example of this phenomenon, if a die is ordered based on a solid/surface model that has not yet been updated, during the assembly process, parts can collide with each other, so that another die must be ordered based on the updated solid/surface model.
In order to automatically update data in a database in accordance with the generation of or a change in a solid/surface model, two solid/surface models must be compared with each other and any difference between them must be detected. While there are several conventional methods for automatically performing or supporting a detection process, with all of these methods, problems have arisen.
As one method, models to be compared are superimposed on a display by using a three-dimensional shape editing function, such as a machine CAD system, and detection of a difference is supported visually. In this case, using a method for employing a common z-buffer for rendering, an interference stripe occurs on superimposed faces, so that when interference strips are used it is difficult to identity a slight difference. Furthermore, since this method is generally carried out by a display engine using z-buffer hardware, it is also difficult for a difference to be automatically extracted using software.
A difference in the two models can be theoretically found by using a ray-tracing function. However, the cost for one rendering performance is high, and to automatically extract the difference between models, for superimposed models, rendering must be performed in various directions, so that there is considerable deterioration of processing efficiency.
As another method, patch division is searched for by using topological information for models generated by the machine CAD system, and a difference in the two models is obtained. However, in actuality, patch subdivision is used to define a shape that occurs when a model is modified, when a pertinent model is divided in order to concentrate on the processing for a model, or when format conversion is performed. In addition, since such an operation would be based on trial and error results provided by the performance of many procedures, it would be impossible to acquire, from among the patch divisions performed to obtain a final model, a division accompanied by a final shape change merely by performing an examination to determine the presence/absence of patch division.
Furthermore, if during the division process entries are not recorded in a log, complicated graphs must be compared to detect the patches that have been divided and to correlate patches before and after their division. Thus, the calculation costs would be increased.
Further, to compare a set of patches that, like IGES (Initial Graphic Exchange Specification), which is one of the file formats used for the exchange of graphic data by different machine CAD systems, are defined at random could, in the worst case, require that a round robin comparison of patches and patch combinations be performed, and the enormous number of procedures that this would entail would render the method unrealistic.
Since an extremely complicated operation is also required to maintain an operating log, this too is not realistic. That is, since as is described above an actual trial and error operation requires the performance of many procedures, a huge number of operating log entries would be accumulated, even for a small change. Further, to carry out the above method, a machine CAD system would have to be changed, and when multiple models derived from the same model are compared, comparisons with the original model would also have to be made.
Another method for detecting a difference in models involves the performance of calculations for a set. However, for this, a high cost is incurred for the calculation of a set of differences for complicated shapes, and in some cases, such calculations can not be completed within a promised time.
Further, for a product such as a nonstandard part that is not easy to be systematically managed, the updating of a database and the adjustment of data tend to be delayed, and design data produced in the past can not effect
Herzberg Louis P.
Hilten John S.
Vo Hien
LandOfFree
Shape analysis, difference detection, and search for... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Shape analysis, difference detection, and search for..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Shape analysis, difference detection, and search for... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3025666