3 posts / 0 new
Last post

It would be neat to have something visual to display tracking of the position of the the yetiborgs around the track. I'm curious to how you guys are managing it right now, and what software you're using for switching between cameras on output.

Any chance you could give a brief outline of the video setup you guys have done?

piborg's picture

What can I say, we currently use some highly evolved wet-ware to drive the cameras, it is called Tim :)

As for the stream selection we have a bit of custom code for taking in all of the camera RTSP feeds and showing them on screen in real-time.
The operator then has a keypad to select one of these, which is fed back out as an MJPEG feed to the machine Claire does her commentary from.
The commentary machine has a simple bit of code which reads in the MJPEG stream and displays it in full screen to one of the monitors.
We can then put videos and such in front of that output to insert them into the feed.
As this is all in real time the machine to do this is reasonably powerful.

For the actual streaming we are using Open Broadcaster Software (OBS) and capturing a full monitor output.
The machine has a second monitor from which Claire can read the chat messages, select videos to play and so on.
This machine is a lot more standard, but it does have an older Nvidia card fitted as the Intel chip was struggling to both record and run the spin view at the same time :)
As a side note we have started making use of OBS as a screen recorder as well, it does the job rather well.

We would love to have some proper display tracking of the robots, unfortunately there is really only the cameras to work that out from.

We are intending to put several light gates into the track so we can get split times for the robots.
These could be used for auto-switching cameras, the trouble is they cannot really tell where the action is.

Have you guys looked into

Have you guys looked into position tracking using Bluetooth beacons? They could be attached to the yeti's and would only have a minimal weight additions, plus lack of pulling processing power. A couple monitor pi's and some triangulation, I think you could get something reasonably accurate to monitor them.

I saw openBeacon, which might be worth a look if you want to research that way.

I'm still going to revisit that rtsp stuff...

Add new comment

Filtered HTML

  • Web page addresses and e-mail addresses turn into links automatically.
  • Allowed HTML tags: <a> <em> <strong> <cite> <blockquote> <code> <ul> <ol> <li> <dl> <dt> <dd> <pre>
  • Syntax highlight code surrounded by the <pre class="brush: lang">...</pre> tags, where lang is one of the following language brushes: bash, cpp, perl, python.
  • Lines and paragraphs break automatically.

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.
Comment Images
Files must be less than 10 MB.
Allowed file types: png gif jpg jpeg.
Comment Attachments
Files must be less than 10 MB.
Allowed file types: txt pdf nfo doc docx rtf jpg png gif bmp zip tar gz csv xls.
This question is for testing whether or not you are a human visitor and to prevent automated spam submissions.