Keep in mind that with something like the limelight, you’re not just paying for the hardware. That is only one piece of the puzzle. The other, rather large piece, is software. While the hardware in this OAK camera may be lightyears ahead of the limelight*, the essentially plug-N-play software certainly isn’t. The primary reason the teams I have been on use the limelight, is because of how unbelievably easy it is to setup a robust pipeline that “just works”.
Look back to the pre-limelight era; Vision was an extremely lofty goal that many teams spent several seasons perfecting/working up to. Nowadays, any team with a limelight or glowworm is basically guaranteed to have functioning vision on their robot. So for teams (like mine) who don’t have an overly strong/developed programming team, widespread adoption of a camera such as the OAK is rather unlikely
I’m certain there are some teams out there that could work some serious magic with this hardware, but for the average team (and without the plug-N-play software), cameras like this would likely end up being an expensive paperweight.
*I haven’t looked at the actual specs (nor really know what to look at/compare)
We have an OAK-D camera available are interested in having it work directly with our RoboRIO. Curious if anyone has been able to compile and run the Luxonis samples directly on the RoboRIO with the OAK-D camera connected through the USB? So far we haven’t been able to figure out the dependencies.
The success of limelight and retro reflective tape is almost a double edge sword.
It’s made things almost too easy where it doesn’t provide a problem that inspires much thought.
I’d argue this warrants an entirely different thread of “COTS is ruining X Y Z”. There are quite a few of these threads already. In regards to the original question, @cad321 sums it up perfectly.
FRC is mostly a mechanical competition (when comparing robot v. robot). There is not much incentive for software improvement, except in the situations to accomplish mechanical impossibilities (or effectively so).
Hard disagree. Every year, I have at least one student who is part of another subteam who wants to “try” a little bit of programming. There is so few things that don’t require a decent background in Java and no mentor shoulder-coding with the kid. The limelight fixes that. I’ve had multiple business students feel like they’ve been able to make a significant impact on the robot because they “solved” vision with the limelight. It was hard for them, but they figured out how to make something work in the scary realm of computer science. That’s true inspiration.
