Method for the perspective visualisation of changeable shapes.

A perspective screen display of arbitrarily shaped bodies, as is common today in so-called CAD systems, can not be realized mit.vertretbarem effort so far in numerically controlled machine tools or programming stations. Generally, we limit ourselves here to the two-dimensional representation of various views of a workpiece.

As recent publications, for example, EP-A2-0 089 561 and EP-A2-0 089 562 may be mentioned here.

Three-dimensional images of the work piece as "transparent" wireframe, such as those offering the control Mazatrol T-1 of the company Yamazaki, are very quickly become confusing because edits are represented only by the display of the tool center point path.

The prior art methods of representation are not satisfactory, to be especially when using the graphical representation via check which program has been in the programming of numerical control and corrected if necessary.

The invention has for its object to improve the graphical representation on screens, and three-dimensional-structure in parallel projection spatially represent to ermöglichen- without the previously necessary computational and instrumental application rates on the screen, which by lying in the foreground image parts concealed edge and surface pieces remain invisible.

This object is achieved by a process whose steps are shown in the characterizing part of claim 1.

The particular advantages lie in the inventive process is that the image of any complicated workpieces produced by the entry of the necessary for their preparation tool movements. Thus, one can follow the action of each processing step on the workpiece directly on the screen.

The necessary procedures for the calculations can be performed with all common microprocessors. The memory required depends on the size and shape of the workpiece. To carry out the procedure for the 3D display, however, only 1K bytes are needed to space, so that on application of the process is also used in small microprocessor systems in question. The drastic reduction in the need for 3D representations effort is possible through systematic replacement of complicated stereometric calculations by simple one-bit operation. Instead numerical intersection calculations the information standing in the pixel memory is to be evaluated directly.

Referring to the drawings will be explained in more detail with the aid of an embodiment of the method.

Show it<ul><li>1 shows a workpiece with indicated average levels;</li><li>Figure 2 an end mill shown in part with indicated average levels;</li><li>3a overlays of different to e sections of the workpiece and the end mill;</li><li>4a overlays of different</li><li>to c-sections of the modified workpiece in the pixel memory;</li><li>5a representation wrought</li><li>to c workpiece at different stages based resulting outer contours;</li><li>Figure 6 is a flowchart of an algorithm for determining contours;</li><li>Figure 7 is a schematic diagram for determining an outer contour at two diagonally staggered sectional planes;</li><li>Figure 8 is a schematic diagram for determining an inner contour at two diagonally staggered sectional planes;</li><li>Figure 9 shows a block diagram.</li></ul>

A three-dimensional Cartesian coordinate system is-according to Figure 1 arranged as follows:<ul><li>The front bottom left corner location of a workpiece W is the point of origin X and Z axes point in the image plane to the right or above. The Y axis has backwards into the image space. Each workpiece W is stored internally in shape of the corner of coordinates perpendicular to the Y-axis arranged sections Yo to Yn. The distance between the individual cutting planes is constant. The number of cuts Yo to Yn depends. on the size of the workpiece W and the maximum resolution of available stehenlen graphics part.</li></ul>

The representation of a section Yi is effected by the X and Z-coordinate origin of a corner point and by the X- and Z-vectors of the further corner points in each case relative to the previous point.

In the same coordinate system can be in accordance with Figure 2 coordinates of cuts Fo calculated to Fn a tool F. The tool, for example an end mill F, also can be represented by sections Fo to Fn. is to determine their coordinates, with the cutter radius as parameters per section Fi calculate once a sine function. It is of course not limited to end mills. It can also store the cuts form cutters.

The coordinate values ​​of the workpiece sections Yo to Yn are stored in a memory section M, which may be implemented as a RAM.

The coordinates of the cuts Fo to Fn of the tool F may also be stored in a RAM that is used as a storage section M '.

In Figure 3, the overlapping of the slices shown Y1 W simplified to Yn-1 of the workpiece with the sections F1 to Fn-1 of the end mill F in different stages. The overlay takes in a pixel memory E1 instead, which contains all image information in the form of picture elements, called pixels, the contents will not be displayed directly on the screen.

The superposition of the sections of the mill Fi F and cuts Yi of the workpiece W is carried out, according to the numerical controller, which determines the tool path or tool path according to their program. the overlapping of cuts Yi and Fi are changed in a predetermined way this data accordingly.

After the initial values ​​of the center point (X, Y) and height (Z) of the tool F are known, one begins with the foremost section of the workpiece W (Y1) which is changed by the tool F, in the non-displayed Bildpurikt memory write .E1. After that transmits the section F1 of the tool F, which corresponds to the Y value Y1, also in the pixel memory E1, determines the inner contour of the image, and replaces the old section coordinates of the workpiece W by those of the modified inner contour Y1 '. This process is repeated until all of which are changed in question sections Yi '.

If it is in the processing step - as in example shown - only to a vertical insertion without lateral movement of the cutter center point, so can the workpiece W, as described further below, to display them. In general, the tool or workpiece F moved west but on arbitrary paths in space. In this case, the change of the intersection coordinates must be carried out anew for each tool position. Appropriately, each section Y edited one = Yn of the workpiece W only once per NC block, by transferring all tool cuts defined by the NC block for the value Y = Yn sequentially in the pixel memory E1 and determined only at the end, the inner contour, when all processing operations for the relevant section Yi have been completed.

The section coordinates the modified workpiece sections Yo 'to Yn' be saved as a new section coordinates average memory M.

To build a perspective image of the machined workpiece W to Yn 'in the pixel memory' the modified sections are Yo 'E1 enrolled. This method sections are illustrated in Figures 4a-c and 5a-c.

First E1 is shown in FIG 4a, the foremost Werkstüokschnitt Yo 'in the pixel memory enrolled and it is determined a resulting outer contour Ar.

Since in the chosen example, the router F has the foremost section Yo not processed, the contour of the "edited" section Yo 'is the same as the original contour. The resulting outer contour Ar is shown in FIG 5a on the screen.

In the second sectional plane Y1 'is a removal of material takes place, and the new contour is written from the memory section in the M pixel memory E1. The new contour of the section Y1 is offset 'by a few pixels in the X and Z direction relative to the front section Yo', which is shown in Figure 4b.

In pixel memory E1 is based on the now inscribed Image information determines the resulting outer contour Ar shown in FIG 5b in addition to the screen.

The following are the other sections Y2 'to Yn' consecutively enrolled in the manner described in the pixel memory E1, where after each Schnitteinspeicherung determined each resulting outer contour Ar based on the image information and then displayed.

4c shows the final stage is shown in which all the cuts Yo 'to Yn' are fed to the image point memory E1. The values ​​determined by this image information resulting outer contours Ar are shown in Figure 5c. This representation corresponds to the image of the processed workpiece W ', as will be made on the numerical control of the operator screen in front of eyes.

The concealed by lying in the foreground cuts parts later lying sections are not automatically transferred to the display because they are within the previously determined outline Ar.

By changing the offset of the cuts Yo 'to Yn' against each other, the viewing angle can be varied. So are oblique image views from different sides possible. Return to the order of the mean processing in a way, producing images of the rear of the workpiece W '. Swapping the incremental coordinate values ​​within each section (for example, X =_Z; Z = X)., Can be to + 90 ° rotated on their sides or rotated by 180 ° with the bottom pointing upwards represent the workpiece W '. The picture sliced ​​work pieces W is possible that you start the 3D display only at a section located in the middle.

Additional possibilities for variation with color graphics. Thus, top and side surfaces easily be displayed in different colors. A dimension of the workpiece W 'can Y10 by color highlighting for example every tenth section' to be clarified.

Finally, one can introduce a zoom function by W treated from the outset a magnified image of the workpiece in the manner described above.

It is understood that a different display device (plotter or the like.) Can be provided instead of a screen, and that the numerical control can also be replaced by a programming station or the like..

The above-described determination of the internal contours of the altered sections Y 'and of the resulting outer contours Ar is carried out by a simple algorithm whose flow chart is shown in FIG. 6 This algorithm described in sufficient manner for the skilled worker by the flowchart is used to update the average coordinates for a tool movement.

It is characteristic of the simplicity of the process, that only one-bit operations must be performed with the algorithm.

Since any arbitrary curved line can be approximated by a staircase line, intersection points of circles, straight lines, ellipses etc. to determine in this way.

The algorithm of Figure 6 is used to determine the polygon an outer or inner contour of a built-up of any number of horizontal and vertical lines of the image.

The example shown in Figure 7 for determining an outer contour is an image that consists of two diagonally staggered rectangles same Dimension. This image can be written by the vector generator of a graphic part in a non-displayed on the monitor level of a pixel memory. As an initial parameter, the coordinate values ​​of any point be on the contour to be determined X_O , Z_O and the direction of an adjacent contour point relative to the starting point needed. In the example, the lower left corner of the image as the starting point X_O= 3, Z_O= 3 defined. The neighbor point lies in the positive Z direction. Using a test routine according to the algorithm of Figure 6 will now be determined whether the left-from the start point pixel X = 2, Z = 3 is set or cleared. From this information you then the coordinates of the next test image point X = 3, Z = 4, and so gets away. Figure 7 shows the sequence of the test steps is shown as a sawtooth line. A corner of the contour is reached when the AX after a jump towards the top right = 1, △ Z = 1 tested pixel is deleted (outside corner) or if the sampled after a left jump △ X = -1, △ Z = 0 pixel set (inside corner). If one of the two cases, so the coordinates of the corner section in memory M are stored, carried out a rotation of the coordinate system through 90 ° and the test routine continued until the start point is reached again. As Figure 7 shows, one obtains the vertex coordinates of the outer contour (3,3), (3,11), (10,11), (10,16), (22,16), (22.8), (15 , 8), (15,3), (3,3) or in the form of incremental 3,3,8,7,5,12, -8, -7, -5, -12, at which by the number of increments is shown the direction arrows.

If one selects as the starting point X_O = 4, Z_O = 3, and the neighboring point in the positive X direction X = 5, Z = 3, is obtained with the same algorithm, the coordinates of the inner contour, as shown in FIG. 8

also here is the representation of the coordinates selected in incremental form (11.5, _-5,3, -7, -8), so a maximum space utilization, raschester screen layout and flexibility in the choice of the viewer's observation point is ensured.

The block diagram shown in Figure 9 illustrates a device for performing the method of the invention. An alternative, control computer RS ​​and graphic calculators RG digital computer R is connected via data lines D with a memory MA for the contours algorithm, which is designed as EPROM. Via data lines D of the digital computer R is connected to Fn with the configured as RAM storage section M, M 'for the coordinates of the workpiece sections Yo to Yn and the tool cuts Fo.

Other data lines D connect a so-called video controller VC with the digital computer R.

The video controller VC in turn is connected via other data lines D 'with an image point memory E1, whose image information is not displayed.

A second pixel memory E2 containing the information required to display image information transmitted from the video controller VC via data lines D '' from the pixel memory E2 to a screen CRT and displayed there.