coordinate transformation - robot

Hello master:

as I knew, unity of the coordinates of the Center back by "Match" is 'pixcel' in the picture. but in fact, we can send coordinated by unit "mm" or "cm" to the robot application. My question is how transorm the coordinate "pixcel" "mm" or "cm". y at - it that no master could give me information or for example on how to do it?

Thank you much in advance.

Thank you and best regards

viming. Chen

Tags: NI Hardware

Similar Questions

SYSTEM of COORDINATES TRANSFORMATION (I think)
Hi all

I need help here. I have a table (which started as a shape file) with a different coordinate system than I use. I can change the SRID and diminfo easy enough but the x / y coordinates are in a different system and I do not know how to convert them.

Help, please!

Here are the details:

-An example of the geometry column I'm shooting for registration:
```
MDSYS.SDO_GEOMETRY(2001, 8265, NULL, MDSYS.SDO_ELEM_INFO_ARRAY(1,1,1), 
MDSYS.SDO_ORDINATE_ARRAY(-79.249266,35.774167))
```
-An example of record of the column of my table of problem geometry:
```
MDSYS.SDO_GEOMETRY(2001, 8265, NULL, MDSYS.SDO_ELEM_INFO_ARRAY(1,1,1), 
MDSYS.SDO_ORDINATE_ARRAY(2012815.139204471,768438.258117677)) 
```
I already changed the SRID in the latter, but the real problem I have is in the MDSYS. SDO_ORDINATE_ARRAY. How can I convert my coordinates of problem (the example of the latter) to the same kind of information as the previous example for each record in the geometry column?

Thank you in advance!
Indeed, for this, you'll need transform geometries of one system of CS (your form source) to another (existing table used).

It's little script that allows you to transform your 'table' or 'sex', based on the

SDO_CS. TRANSFORM_LAYER

[http://download.oracle.com/docs/html/B14255_01/sdo_cs_ref.htm#sthref1408]

Change in this script SOURCE_TABLE with your TABLE NAME
Change in this script TARGET_SRID with your TARGET SRID
Replace in the SOURCE_TABLE_SPATIAL_INDEX_NAME script with the name of the spatial index on your table in the source
--===============================================================================
-Run SDO_CS. TRANSFORM the layer
RUN SDO_CS. TRANSFORM_LAYER ('SOURCE_TABLE', 'GEOMETRY', 'SOURCE_TABLE_TRANSF', TARGET_SRID);
Quickly transformed layer;

--===============================================================================
-DROP SOURCE TABEL SPATIAL INDEX
DROP INDEX SOURCE_TABLE_SPATIAL_INDEX_NAME;
COMMIT;
--===============================================================================
-execution of the procedure to update the source table
DECLARE

CURSOR c1_cur IS SELECT sdo_rowid, geometry OF SOURCE_TABLE_TRANSF;
l_tekst varchar2 (20);
c_rowprocessed CONSTANT number (3): = 500;
my_loops pls_integer;

BEGIN
my_loops: = 0;
FOR c1_rec IN c1_cur LOOP
UPDATE of SOURCE_TABLE one
SET a.GEOMETRY = c1_rec. geometry
WHERE a.rowid = c1_rec.sdo_rowid;

my_loops: = my_loops + 1;
If mod (number of lines c1_cur %, c_rowprocessed) = 0 then

commit;

end if;

end loop;

commit;

end;
/
Geometry of layer QUICK update;
--===============================================================================
-REMOVE the Source heading in USER_SDO_GEOM_METADATA REMOVE USER_SDO_GEOM_METADATA
WHERE
TABLE_NAME = 'SOURCE_TABLE.
AND
COLUMN_NAME = 'GEOMETRY ';
COMMIT;
--===============================================================================
-integration with the DETAILS DIM_ELEMENT and SRID USER_SDO_GEOM_METADATA
INSERT IN USER_SDO_GEOM_METADATA
VALUES
(
"SOURCE_TABLE."
"GEOMETRY."
MDSYS. SDO_DIM_ARRAY
(
MDSYS. SDO_DIM_ELEMENT ('X', 0, 300000, 0.00001),
MDSYS. SDO_DIM_ELEMENT ('Y', 0, 300000, 0.00001)
),
TARGET_SRID
);
COMMIT;
--===============================================================================
-Recreate the index
CREATING INDEXES
SOURCE_TABLE_SPATIAL_INDEX_NAME
WE
SOURCE_TABLE (GEOMETRY)
INDEXTYPE IS MDSYS. SPATIAL_INDEX;
Commit;
INVITES the 2 Spatial Index created;
--===============================================================================
-DROP the SOURCE_TABLE_TRANSF table
DROP TABLE SOURCE_TABLE_TRANSF;
COMMIT;

Hoping that it would be.

Luke

How to make 'Calibration Simple' with 'Vision calibration Training' or "Wizard of Vision"

Hello master:

I want to do 'Calibration Simple' with 'Vision calibration Training' or "Assistant vision", I want a model calibrated using this way of exit? How to do?

I had already read "Simplate Calibration.vi".

Thank you.

https://forums.NI.com/T5/machine-vision/coordinate-transform-robot/TD-p/3287696/page/3

Choose and place using labview and or vision acquisition

Hello world

I'm doing a project studying on Vision guided pick and place of a robot (abb) industrial. I would like to know the steps involved in the creation of the block.

I locate the object, move his webcam cooordinates. Then made a pattern match, and would send the cooordinates to the microcontroller. then from microcontroller for control of robot... then the industrial robot should choose the object and place it in a predefined area...

I would be extremely grateful if you guys can help me because I am new to LabView.

Thank you

Pradeep.M

([email protected])

What you describe is quite complex, but here are a few tips. The key is to establish a correlation between the coordinate system of the robot to the coordinate system of the camera. I guess that the camera is statically located above the pick-up area? I move the robot at each corner of the frame to its choice position vertically and note the position of the robot at these locations. These 4 points in space will be correlated to X, coordinates of pixels in the camera image. Basically, you need to write a sub - VI with entries being pixel X and is coordinated and coordinates output being the robot.

Writing a test application saying the robot to get pixel location to any X, Y in the framework to test your Subvi. If this does not work, then you need to set up a correspondence to the model. You probably want to do a geometric pattern match. Take a look at this example: http://zone.ni.com/devzone/cda/epd/p/id/5555

You will need your pattern match algorithm to return both the coordinates for your robot, and the orientation of the tool needed for good pick up the object (if the pick-and-place robot tool requires to be in a specific direction). If it's basically up to you will convert the object X, Y and rotation angle in the framework that you receive correspondence from model to any coordinate system, the robot uses.

The placement algorithm could be simply an adjustment of orientation to the object being investment and then investment positions could be an array of coordinates of robot which you browse after each pick.

Be sure to implement security mechanisms in your algorithms so that the robot can never go somewhere outside of a safe range of coordinates.

Transform the coordinates Y wrong

In the area of transformation the Y coordinates are false. The object is above the zero of the art Council series still coordinate is reading a negative number. I changed to Global leaders, but it's always reading a negative number. Help?

pvanderh, as a result of what Larry said:

To change the origin of the rule in CS5 are the same as in CS4 and earlier versions, you can:

0) close down Illy (you can change the closed preferences with Illy file);

(1) find and open the AIPrefs (Win talk) or Adobe Illustrator Prefs (Mac talk) of the file. It is a hidden file in the folder of Adobe Illustrator CS5 Settings;

(2): find and replace the following two bits of code:

/isRulerOriginTopLeft 1 > /isRulerOriginTopLeft 0 (replace the 1 with a 0)

/isRulerIn4thQuad 1 > /isRulerIn4thQuad 0 (replace the 1 with a 0)

This is a global change.

Transform a user created the system of coordinates to Oracle supplied one
Hi all

I'm looking to transform an obsolete network National British Ordnance Survey projection, so that the data that contains the obsolete projection can be queried using coordinates gird National British.

I think I need to insert an entry in the table SDO_PREFERRED_OPS_USER and not SDO_PREFERRED_OPS_SYSTEM. However I do not know what value to use for the COORD_OP_ID. Is the COORD_OP_ID value related to the SOURCE_SRID, which is the obsolete projection; the value COORD_OP_ID for the TARGET_SRID or a new value?

I I take then insert another line with the SOURCE and the TARGET of the swapped SRID round so that changes may occur in the sense the other way around.

Kind regards

Tim
Generally, you don't need any rules of transformation between two screenings. The system automatically performs the reverse projection, datum transformation and projection forward then. In addition, even if a projected SRID is now considered to be obsolete, this does not change the fact that the system can determine a default transformation approach. For most applications, the default value is right.

Do you need a rule, however, if the default processing line is not what you prefer. For example, if you want to convert between NAD27 and NAD83 using NADCON and not the transformation of default reference, you need an appropriate rule. If this is the case for you, please let me know and I can help you with the rule.

Regarding the choice between SDO_PREFERRED_OPS_USER and SDO_PREFERRED_OPS_SYSTEM: If you use the system, it will become the default for all users of space in your DB. If you must use a string of personal transformation that you don't think that other users of your database would agree with, then SDO_PREFERRED_OPS_USER is appropriate. The user is a tag of your choice VARCHAR2, not a database user. Other users of the database can share your setting, if they use the same tag.

SOURCE_SRID, TARGET_SRID and OP_ID should be consistent. If in doubt, please let me know your specific processing needs, and I can provide you with a rule.

For your specific example, I need more in detail to determine if you need a rule, at all. It seems that you can not. Please let me know source & target SRID, and why you think you need a default transformation.

Thank you

Mike

Graphic intensity scale and the Transformation of coordinates

Hi all. First post ever so please bear with me. I'm trying to plot the data that I am buying a load cell in coordination with the X - Y position, I get my motion control system. The news of load cell must be the z axis. I'm receiving the order of movement data, then acquire the load cell signal on average and store them in a table. I would like to plot these data for each movement of x and there axis so that I can see a real-time mapping of the data that I have gained. I joined a program that computes XY positioning required with load cell data simulated by a random number. Drawing the x - y graph coordinates XY is the standard path measurement I use. I wish that the graph of the intensity to draw the same way with the correct details used on the axes. If I could put the xy graph and the graph of the intensity, it would be a major bonus. I can't wait to hear any response that helps me on this one. Thank you in advance for your time and effort. Only, I'm stumped on this one!

Keith

I'm sorry. Realized that I do not understand a Subvi. Deleted so the program should work if download you it directly. Just a few notes: Spray size in inches and normally between 2 and 20 inches, degrees lateral spray angle is normally between 10 to 45 degrees. Angle of transverse jet is normally between 2 to 10 degrees. Pin diameter is in inches and usually falls 0, 020-0, 080 range.

Transformation of the coordinate system
I use company oracle 10.2.0.1

I'm trying to convert our ArcSDE that sit on top of RDBM oracle database, which is used to visualize on the map of the Earth to the oracle database. I have elevation data, latitude and longitude for a region (Rectangle shape) I want to insert into a column of type GEOM. I was going through the spatial reference guide, which has lots of information. My question is can I use this stmt to insert data of a shape of rectangle with the longitude/Latitude data?

INSERT INTO (VALUES) target_point
1,
"cola_a,"
MDSYS. SDO_GEOMETRY)
2003 - polygon 2 Dimensions
NULL,
NULL,
MDSYS. SDO_ELEM_INFO_ARRAY (1,1003,3), - a rectangle (1003 = outside)
MDSYS. SDO_ORDINATE_ARRAY (1, 1, 5, 7) - only 2 points needed to
-define the rectangle (bottom left and top right) with
-Cartesian-coordinate data
)
);

I also want to know what is the difference between that and the following stmt, that store all vertices?

INSERT INTO (VALUES) cola_markets_cs
1,
"cola_a,"
SDO_GEOMETRY)
2003 - two-dimensional polygon
8307,-SRID for ' Longitude / Latitude (WGS84) "coordinate system".
NULL,
SDO_ELEM_INFO_ARRAY (1,1003,1), - polygon
SDO_ORDINATE_ARRAY (1.1-5.1, 5.7, 1.7, 1.1) - all vertices must
-be defined for rectangle with geodetic data.
)
);

I'm new to spatial database and try to learn. Appreciate any help.

Thank you
Since you have rectangles, you have two options.

Create a rectangle of 5 points (you will need to repeat the last coordinate to close the polygon).
Also affect the SRID 8307 since the values long/lat.

SDO_GEOMETRY (2003, 8307, NULL,)
MDSYS. SDO_ELEM_INFO_ARRAY (1,1003,1),
MDSYS. SDO_ORDINATE_ARRAY (long1 lat1, long2, lat2, long3, lat3, long4, lat4, long1, lat1))

And you can then use SDO_GEOM.sdo_centroid to calculate the centre of gravity in the application.

Regarding the altitude, is there an elevation for each coordinate value?
10.2, we do a lot with the values of height, but you can store around the geometry.

Then eventually you create geometry 3D like this:

SDO_GEOMETRY (3003, 8307, NULL,
MDSYS. SDO_ELEM_INFO_ARRAY (1,1003,1),
MDSYS. SDO_ORDINATE_ARRAY (long1 lat1, alt1, long2, lat2, alt2, long3, lat3, alt3 long4, lat4, alt4, alt1, lat1, long1))

But in this way, everything becomes a little more complicated. So if you just want to make altitude autour, you better
account in a separate as well as the geometry column column.

Siva

Distance between 2 points in geo coordinates

Everyone knows hot to calculate the distance between 2 points on the surface of the Earth, given their contact details?

(for example: 1=30.45N/43.23E and 2=35.11N/19.05W)

A program for HP50G or 42, or 41 c would be fine, a formula or algorithm as well.

Hello andy11

Forms are to is supported on a perfect sphere (which is not true for the Earth, but it makes it easier):

The way is to calculate the secant between two points and then calculate the length of the arc between the two points.

LA1/2 = Latitude of the point 1/2

LO1/2 = Longitude of the point 1/2

Transform in x, y, z coordinates on the unit sphere:

X 1/2 = cos(La1/2) * cos(Lo1/2)

Y1/2 = cos(La1/2) * sin(Lo1/2)

Z1/2 = sin(La1/2)

Calculate secant length s:

s = ((x2-x1) ^ 2 + (y2 - y1) ^ 2 + (z2 - z1) ^ 2) ^ 0.5

Half of the s is a sin of angle between the points:

distance on the sphere:

d = RE * 2 * arcsin(s/2)

That's it, easy to transfer this for RPL. The results are of course proven RE RADIUS, so results may differ a bit.

Greetings

calcpeace

Coordinates of the cursor & Rectangle coordinates does not not to

Hello!

I actually paint a drawing a rectangle on xy graph using the mouse by clicking on the events and I need the contact information exactly the way it is extracted by cursor. The indicator of "coord2" & "cursor 0" should give me the same results. I need output of coord 2 as cursor 0 hope I managed to pass my point. Maybe I need to manipulate with scale. Check the attachment.

Sets,

I have attached a VI that will transform your mouse up and down the coordinates which are by default in pixels for the x - y coordinates. I also noticed that you could not make a rectangle when you started at the top left and ended at the bottom right, so I added a logic, so you can start and finish anywhere and have the rectangle belong.

I also noticed that when you draw the rectangle that the corners are drawn a little below and to the left of the cursor, that is the question you are trying really hard? I just looked in that a bit but could not understand why that was happening.

Circle quick algorithm (Bresenham?) drawing to transform circular Hough

I worked on a circular Hough transformation algorithm for circle detection and I am interested in finding ways to make a little more efficient. At high level, I implement some queueing of client/producer model between the data collection and analysis/filtering, but I think that the circle of operation for the accumulation of Hough of drawing could use some breaking in. That's where I'd love help from you, the experienced programmers of LabVIEW.

I used simple trigonometry draw pixels with some angular resolution, but I think using a circle drawing algorithm as the algorithm of circle in the middle (variation of Bresenham circle) is probably smarter in the end. The abomination of hidious of LabVIEW script I created for her, however, is not complete as quickly as I had hoped. I followed the C code written here: http://members.chello.at/~easyfilter/bresenham.html

You can find the attached VI. It takes in a matrix of pixels in an image binary (0 to 255), with the edges and draw circles centered on each pixel 255 with the given radius. "Drawn" circles are simply a list of coordinates, the output array. Anyone have any suggestions on how to improve this VI? Or maybe even some input to the notion of transformation of Hough set?

Hi Vekkuli!

I have, but a few recommendations for this function. Mind you, this is just my point of view, there could be more advantageous alternatives.
1. For the calculation of the pixel when you used the formula nodes. I think it's a little overdosed because it's only in addition/subtraction, so you can just do that in LabVIEW.
2. When collecting the coordinates, try to avoid using registers at offset and table build to add new items in each iteration. Because arrays are dynamic data types, this causes the compiler to allocate a new, bigger memory, copy the entire table with the new elements, and then free the old memory. Do what each iteration is quite a Devourer of performance. A better approach would be to use indexing or concatenation of tunnels.
3. Borrowing tunnels of indexing to collect data also added, advantage to the outer loop (the iterate over all pixels) paralellizable, which could be a big performance increase, assuming a multi-core PC.
4. You can get a possible increase in speed by drawing octants instead of quadrants, creating 8 points at each iteration.
I also tried my hand to implement this algorithm that creates a circle, here is the result:

On the transformation of Hough, there are examples of LabVIEW implementation, made at the University of Texas. You will find them here.

I hope this helps.

Kind regards:

Andrew Valko

NOR Hungary

Transform TF101 does not start

My ASUS Transformer TF101 eee Pad will not start. When I have it turn on the start screen comes on and turns off in about 5 seconds. Once I do the power button has no effect and the start screen continues to do so until the battery dies. I tried the soft reset button has no effect. I tried to do the hard reset (power at the same time the button and the button low volume) and the display shows the robot Android lying on it comes back with a symbol of a triangle on a panel open on his chest.

Is there anything else I can try before you take or send it somewhere?

Hello

It is the ASUS material with an Android operating system.

Check with the ASUS support and their forums.

ASUS TF-101
http://EEE.ASUS.com/en/transformer-Prime/features

ASUStek support
http://support.ASUS.com/

ASUStek Forums
http://VIP.ASUS.com/forum/default.aspx?SLanguage=en-us

ASUStek drivers
http://support.ASUS.com/download/download.aspx?SLanguage=en-us

I hope this helps.

Rob Brown - Microsoft MVP<- profile="" -="" windows="" expert="" -="" consumer="" :="" bicycle="" -="" mark="" twain="" said="" it="">

CRA geometries give error in CDO_CS. TRANSFORM

Seems that SDO_CS has problems in the transformation of the geometries of the CRA:

-LINE geometry works very well

SELECT SDO_GEOM.VALIDATE_GEOMETRY_WITH_CONTEXT(geom,0.01) val
       ,SDO_CS.TRANSFORM(geom,4326) trafo
FROM
(SELECT sdo_geometry
(2002, 25832, NULL,
   sdo_elem_info_array (1,2,1),
   sdo_ordinate_array(
   500014,5650010,
   500010,5650014,
   500006,5650010)
   ) geom
FROM dual);

TRUE

MDSYS. SDO_GEOMETRY (2002,4326, NULL,

MDSYS. SDO_ELEM_INFO_ARRAY (1,2,1),

MDSYS. () SDO_ORDINATE_ARRAY

9.0001995237432,51.0016646160954,

9.00014251706963,51.0017005861695,

9.00008551017565,51.0016646162345))

-ARC geometry gives ERROR

SELECT SDO_GEOM.VALIDATE_GEOMETRY_WITH_CONTEXT(geom,0.01) val
       ,SDO_CS.TRANSFORM(geom,4326) trafo
FROM
(SELECT sdo_geometry
(2002, 25832, NULL,
   sdo_elem_info_array (1,2,2),
   sdo_ordinate_array(
   500014,5650010,
   500010,5650014,
   500006,5650010)
   ) geom
FROM dual);

ORA-13285

Transformation of the coordinates of geometry error

Cause: A pair of coordinates was outside the valid range for a conversion/projection.

Action: Check that the data to be transformed are consistent with the desired conversion/projection.

The range of coordinates are valid IMO. This bug (?) exists in ORACLE 11.2.0.3 and 12.1.0.1.

Found a workaround to SpatialAdvisor:

http://spatialdbadvisor.com/oracle_spatial_tips_tricks/318/circular-arcs-in-geodetic-polygons

Changed the SDO_GTYPE and the Ansdo_elem_info to MULTIPOINT, transformed and changed their return:

SELECT sdo_geometry(2002,trans.tgeom.sdo_srid, NULL,
         trans.geom.sdo_elem_info,
         trans.tgeom.sdo_ordinates)
  FROM
(SELECT SDO_CS.TRANSFORM(
           sdo_geometry(2005,orig.geom.sdo_srid, NULL,
             sdo_elem_info_array(1,1,sdo_util.getNumVertices(geom)),
             orig.geom.sdo_ordinates),4326) tgeom,
             geom
      FROM
      (SELECT sdo_geometry
      (2002, 25832, NULL,
         sdo_elem_info_array(1,2,2),
         sdo_ordinate_array(
         500014,5650010,
         500010,5650014,
         500006,5650010)
         ) geom
      FROM dual
      ) orig
) trans;