New PixHawk quadcopter board adds visual waypoints

The new PixHawk quadcopter controller board has some serious horsepower onboard! From the team's post:


"Its two Intel Core 2 DUO 1.86 GHz cores run only at 10% CPU usage with our competition software. The system has been mainly designed for future Computer Vision on Micro Air Vehicles research in PIXHAWK at the Institute of Visual Computing of ETH Zurich. We will use it to localize the MAV with pure onboard processing based on natural features. The current working state uses artificial features to localize.

Regarding our competition entry, we now reached a state where our PIXHAWK Cheetah ALPHA system not only hovers autonomously with no wireless video transmission using ARTK / computer vision, but also starts, lands and follows waypoints while detecting objects along the path. We're now optimizing the system to make sure it works fine in the competition itself. Part of these efforts is to build BRAVO, which has an optimized frame and electronics design and thus reduces the all-up weight from 1200g to about 900g."

Views: 1573

Comment by Rana on June 27, 2010 at 11:40am
O' my God ! That's really great Chris !
Comment by Chris McNair on June 27, 2010 at 11:43am
Thats some serious power!
Comment by Overwatch on June 27, 2010 at 11:47am
I wonder what kind of wattage those things get. They look nasty... and good.
Comment by ionut on June 27, 2010 at 11:56am
I'm wondering how much board costs ?
Comment by Nullified on June 28, 2010 at 6:15am
Does anybody know which single board computer they're using?
Comment by Guido Armaris on June 28, 2010 at 6:40am
Kontron board, see here
Comment by george on June 28, 2010 at 9:32am
Could someone explain better what this does? If I understand correctly it would allow your quadcopter to fly towards different waypoints but at the same time visually recognize things on the ground? i.e. fly a grid looking for its landing platform and then land once it recognizes it?

Comment

You need to be a member of DIY Drones to add comments!

Join DIY Drones

© 2019   Created by Chris Anderson.   Powered by

Badges  |  Report an Issue  |  Terms of Service