LabVIEW Vision Example

[s5511]

We are trying to integrate the 2016 LabVIEW vision example into our robot project. How would we go about 'importing' the vision example into our project?

[Mark McLeod]

You can checkout the tutorials to help get started.

From the Getting Started window (when LV is first started or View->Getting Started Window...)
Select Tutorials on the left side.
See Tutorial 7 - Integrating Examples into Robot Code
and
Tutorial 8 - Integrating Vision into Robot Code

It's pretty much a complete replacement of the default Vision Processing.vi
Porting the Robot Global Data.vi contents
Adding the Vision Support Code

[s5511]

So is the roboRIO processing the image on-board, or is the driver station processing the image and sending the data back? We are wondering how it would perform on the field, and it might be a problem if it is sending the image over the FMS to the driver station, and the latency may be too high

[Maximillian]

Originally Posted by s5511

So is the roboRIO processing the image on-board, or is the driver station processing the image and sending the data back? We are wondering how it would perform on the field, and it might be a problem if it is sending the image over the FMS to the driver station, and the latency may be too high

You can do it either way. If you want to process the image on board you copy the example code to the Vision processing VI. If you want to do the image processing on the driver station you can copy the example code to the Dashboard project in the vision processing loop that's already in the project by default.

My team did the processing on the dashboard and sent back the coordinates of the goal to the robot via network tables. Let me know if you want to see our dashboard project and I can send it to you for reference.

[s5511]

Yeah, I'd love to see your code. It would be great if you could send it to cortechsrobotics@gmail.com


We copies the code in the example vision processing vi and just transferred it with its dependencies to our project. So does that mean it's running on-board?

I'm curious as to what are the pros/cons of running it on the dashboard or on board. What made you do it on the dashboard?

Originally Posted by Maximillian

You can do it either way. If you want to process the image on board you copy the example code to the Vision processing VI. If you want to do the image processing on the driver station you can copy the example code to the Dashboard project in the vision processing loop that's already in the project by default.

My team did the processing on the dashboard and sent back the coordinates of the goal to the robot via network tables. Let me know if you want to see our dashboard project and I can send it to you for reference.

[adciv]

We did our processing on the roborio this year. Primary reason is that it was powerful enough for what we needed to do. For the off season we're working on an on robot coprocessor to do the image processing for us.

Our robot code is on GitHub here under 2016-RoboBees-OffSeason:
https://github.com/FRC-836/

You WILL need the 836-Library to be installed in userlib. Documentation is spotty in many places and we're working to improve that. Message me if you have any questions. Also, those are works in progress as we go through the off season so there will be udpates.