Using CatBot 2.0
CatBot 2.0 simplifies your optimization workflow by linking Catia and Robot directly. Without the need to write an intermediary XML file, the process of building and analyzing a Robot model is significantly faster. By building a new Robot model in every analysis loop, you are also able to make elements appear and disappear in your solution, making growth, adaptation or degeneration possible. The CatBot node is now linked to the Catia process via a synchronizer and the syntax for the DOS script should read as follows:
[CatBot executable location] [your working filename] [modeFrontier flag] [incremental save flag] [your working directory]
[modeFrontier copy command] [path to your ouput file] [modeFrontier local process directory variable\output file name]
[exit command] [modeFrontier exit error level variable]
The first line executes the CatBot program taking your working file name, two inputs flags and your working directory as its arguments. The second line copies the ouput file from the working directory, where it is initially created, to the local modeFrontier process directory for reading. The third line exits the node, returning 0-Success or 1-Failure. The syntax of a real CatBot node would look something like this.
E:\AF_Tools\CatBot\CatBot.exe GridExample -mf -s E:\AF_Tool\WorkFlow\081103
copy E:\AF_Tool\WorkFlow\081103\GridExample.txt %APPL_PWD%\GridExample.txt
exit %ERRORLEVEL%
Comments
Self-weight and axial stresses in txt file?
Hi Ian and studio,
I was calculating total weight of structure before and am wondering if this can be added to the new workflow? Also, is there a total axial stress value in the txt file? If these values are already there or if I should be using other values to analyze structure, please let me know. (I am currently using displacement).
Thank you,
Gwen
Working!!!
I've got a very simple file with only 9 points and 4 lines working.
here is the syntax for the CatBot node
-------------------------
New Syntax
C:\SimpleWF\CatBot.exe Simple -mf -s C:\SimpleWF\
Copy C:\SimpleWF\Simple.txt %APPL_PWD%\Simple.txt
exit %ERRORLEVEL%
--------------------------
Old Syntax
C:\SimpleWF\CatBot.exe Simple -u -o C:\SimpleWF\
Copy C:\SimpleWF\Simple.dat %APPL_PWD%\simple.dat
exit %ERRORLEVEL%
I have to admit that the problem wasn't solely the syntax here but also my Catia file. In an effort to create a troubleshooting file quickly I had some problems with my inputs which made the catia file exit quickly before it could build a robot file.
So the moral of the story might be to check to make sure the catia file runs through modeFrontier alone before introducing the CatBot component.
CatiaBot Error
Hi Ian,
I'm getting the same error as Daniel, immediately when I run mF, CatiaBot errors out before it updates the catia model. There's no problem making the initial robot file and txt file or getting modeFrontier to produce feasible designs without the catbot node. This is what I have inside my catbot node:
E:\jhl2142\everything_folder\CatBot.exe Midterm_Terminal33_testJ -mf -s E:\jhl2142\everything_folder
copy E:\jhl2142\everything_folder\MIDTERM_TERMINAL33_testJ.txt %APPL_PWD%\MIDTERM_TERMINAL33_testJ.txt
exit %ERRORLEVEL%
Also, I don't know if this is related or a part of a separate issue, but sometimes when I click throught the CatiaBot errors, i get this warning message in robot "Demo version work mode!!!" (Three exclamations) And am prompted to either "authorization update" or "ok." See image here.
Some measure of success!
Danny and I were thinking that the error was occurring because catbot isn't allowing the catia model to finish updating before it starts writing the list of nodes, bars, points, to the new list. So I set a "sleep 15" as the first line in my mF node, and then watched the catout node being written in the "stdout.dat" file. It shows the nodes being written to the list, but is racing the catia file to see which will update first. If catia updates first it errors out. I managed to get a successfully updated new robot file - with no CatiaBot errors - however, it produced a weird hybrid of the previous model inputs with the new inputs. Then I tried "sleep 100" and it was better, but still showing some previous fixed points and old design geometry. It was only when I bumped the sleep up to 500 that I was ensured that the catia model would update and only then, after waiting 8.33 minutes would catbot make the new robot file.
See images here.
.txt error?
John/Everybody;
I just ran into a similar situation where some where failing and others weren't right when they get through the robot phase. I think it might be the output variable in my case. Since we are now dealing with generative geometry the locations of the output data is likely not in the exact same place every time it generates a .txt file.
so instead of using "Absolute Position" when designating the data-mining rule, I think it should be "Relative Position." Did you already do this?
Sleep?
I think the problem has to do with modeFrontier starting the catia process and the CatBot file at the same time. I can see it trying to open but it never lines up sequentially. should I try sleep? I've played a little with it, but even when it waits long enough for catia to finish deleting my powercopies, catia closes out before the robot process can begin/ or the robot process tries to run before catia is ready...... I've tried it with a much simpler model with no success.
Is it possible to see a sample workflow that is working?