Camera Issues (we have tried practically everything!)

[Team1908]

Our team has been having trouble trying to set up the camera so that it returns images to the driver station. This is what we have done so far (in no particular order):

-Installed the driver station update (correctly, so that the compass indicator is showing)
-Used the Axis Camera Setup program from Labview to configure the camera
-Re-imaged the cRIO for Labview using image v10 and built and deployed code to the cRIO (all vision code is still as it was from the default code)
-Ensured that our camera is functional by viewing the live feed through an internet browser on our computer

Assuming that the default code made provisions for sending images to the driver station, the only two things that I can think of are wrong are:

-Requiring the Labview update
-Using the newer v19 cRIO image

The main questions we would like answered are:

-Would not doing either of those things prevent the camera from sending data to the driver station?
-Could anything that we haven't done besides those two things prevent the camera from sending data to the driver station?

Thanks in advance!

[Joe Ross]

Originally Posted by Team1908

Assuming that the default code made provisions for sending images to the driver station, the only two things that I can think of are wrong are:

-Requiring the Labview update
-Using the newer v19 cRIO image

The main questions we would like answered are:

-Would not doing either of those things prevent the camera from sending data to the driver station?

Absolutely, you need both of those.

[Team1908]

Okay, we'll run both of those and try again.

[Team1908]

Okay, we have run both of those updates (as well as the updates that were released recently) and re-imaged the cRIO to version 20. We successfully receive images wirelessly now from our robot to our driver station. The only problem is that the added process seems to be lagging our robots operations of all of its other components. Is there any way we can use a delay to slow down how many times it gets images so that our robot will function normally?

[Greg McKaskle]

Begin lets you configure how the camera operates. What size, compression, and frame rate is it set to? If you lower the frame rate and possibly the others, I think you'll find a good solution.

Greg McKaskle

[Team1908]

Originally Posted by Greg McKaskle

What size, compression, and frame rate is it set to?

We simply kept the default parameters that were given to us in the code. Any code pertaining to the camera was never changed from the default.

[Greg McKaskle]

The default code is always looking for the target, in all modes. If you don't take advantage of this, you can turn off the button that enables the vision processing. You can also do this programmatically by turning off the global for different game modes, based on a button, etc.

If you want to process for targets, it definitely takes CPU, but shouldn't really affect the way the robot drives. Have you moved the vision code into teleop or something?

Greg McKaskle

[Team1908]

Originally Posted by Greg McKaskle

The default code is always looking for the target, in all modes. If you don't take advantage of this, you can turn off the button that enables the vision processing.

Have you moved the vision code into teleop or something?

Greg McKaskle

We aren't taking advantage of the code for finding the target, but we're not quite sure how to disable vision processing.

And no, we did not move the vision code into teleop. The vision code in our project is exactly identical to the vision code in the default project. None of our programmers dared to click on them.

[Greg McKaskle]

Just checking about the location.

The Vision Processing VI contains a loop that either sleeps or gets and processes an image. That decision is made based on a global. If you set that global you can control the vision processing. The place the global is currently set is in Robot Main.

Note this processing is independent of the vision back to the PC. There is a similar button beneath the dashboard that will enable and disable the dashboard images. Since those are not processed by the cRIO, the overhead to retransmit is really small by comparison.

Greg McKaskle

[Team1908]

Originally Posted by Greg McKaskle

The Vision Processing VI contains a loop that either sleeps or gets and processes an image. That decision is made based on a global. If you set that global you can control the vision processing. The place the global is currently set is in Robot Main.

Note this processing is independent of the vision back to the PC. There is a similar button beneath the dashboard that will enable and disable the dashboard images. Since those are not processed by the cRIO, the overhead to retransmit is really small by comparison.

That's excellent! We're glad that this will (hopefully) de-lag communications between the robot and the driver station without attempting to mess too much with the camera settings. We'll try it this afternoon and see how it works.

[Team1908]

We set the global for enabling the vision processing to false using a constant, and now the lag is gone. Thank you very much for your assistance.