I hate it when the best way to state something is also the punniest. You guys remember the DARPA Grand Challenge, right? That used full-size cars with super-customized navigation systems. If you’re just an engineer trying to design better crash avoidance software, or determine traffic interactions between two models of navigation, it’s not exactly cost-effective to buy a Hummer and outfit it with $10,000 worth of gear. The solution: Robocar Z.
This cute little bugger is actually an extremely advanced standard unit for working on autonomous navigation systems. At $7500 it’s not something people will pick up on a whim, but a research budget can afford a few if it means a chance a fat grant from Toyota or something.
Robocar Z runs Linux and has its own API, and is outfitted with two tiny cameras to provide stereoscopic vision. It looks like there are some more sensors on the front there as well (SONAR and whatnot), but I’m not 100% sure of that. It’s totally self-contained, CPU and all.
With a couple of these you can run experiments with multiple vehicles and not worry about wrecking real cars. Plus, they’re so cute you just want to pick them up and hug them.
[via Gearfuse]
$7500 is a lot cheaper than buying real cars.
I admire people like you who take the time to do all this research and then to share the results with others. Due to time constraints I will have to come back and read all the articles you have linked to.
Thanks for the share.
Most of the cost is in the Hokuyo laser range finder, the little orange-topped cylinder seen in the full picture, under the cameras. They run about $4500. They are really amazing sensors though, 8000 points in a 270 degree arc, reading up to 10 meters with millimeter resolution if i remember correctly.
This is a pretty nice platform for autonomous vehicle testing!