Moderator Comment by Chris Anderson on April 2, 2009 at 12:02pm

$300 for the board alone! I think that's going to be hard to compete with custom autopilot boards, which are smaller, lighter and have more appropriate connectors.

Comment by Matt Fisher on April 2, 2009 at 12:08pm

Totally agree. It really doesn't have all the appropriate connections. To be honest I see myself using a Ardupilot in the air and maybe something like this on the ground. It has VGA out so maybe I could do something nifty with a OSD system.

But I guess for $300.00 I could just go buy a netbook to run as my ground station, decisions, decision. Ah well, still a cool little board in my opinion.

Moderator Comment by wayne garris on April 2, 2009 at 12:39pm

VIA Board
regardless of what others might think boards like these are the future of uavs . i say this because APs of the future will be doing a heck of a lot more then just going from one place to another . think if you could run flight gear in the plane itself . this could bring the possibility of running a full machine vision environment in each autonomous unit . I am not going to spell out here exactly what one could do with more power quit frankly I am getting a little tired of peoples lack of imagination on this subject . if anyone would like ,your more then welcome to go through my blogs and with a little intelligence on could surmise what it is I am referring to.
and BTW this has all the right connections looks even better in that regard then the VIA .thanks Matt for showing this.

Moderator Comment by Chris Anderson on April 2, 2009 at 12:47pm

Wayne,

I totally agree that autopilots will (and are) going to do more in the future. And Moore's Law is going to be bringing this kind of power down in price and size. My point is only that history has shown that dedicated autopilot hardware, as opposed to generic platforms (from this kind of dev board to a PDA), tend to work better for our relatively unique needs. If you've got a robot car, you can use generic computing hardware; for a robot plane on the amateur scale (we're going to be FAA-limited to under 4 lbs), you probably want something more specialized.

Comment by Matt Fisher on April 2, 2009 at 12:53pm

I agree, as this board would be a good starting point to learn about the new pieces of tech embedded in the board, it's still not a 'UAV Board'. So its obviously not optimized for this purpose which means there will be some waste on the board somewhere. I guess it would be smart to look at a board like this and see what are the smartest things to start porting over to a UAV board.

Moderator Comment by wayne garris on April 2, 2009 at 1:04pm

"(we're going to be FAA-limited to under 4 lbs)"
do you mean there talking about limiting diy uavs ? whats your source ? also i think due to Moore's law and the inability for large companies to move quickly well will start seeing giants fall on their face . (and i am sure many people here will help with that ). BTW i think you ardupilot is awesome I'll bet dave @ unav hates you . (unav is a perfect example of the effects of Moore's law). massive cuddos to you and your understanding of Moore's law. keep up the good work !! also any plans for a fly-in this summer?

Moderator Comment by Chris Anderson on April 2, 2009 at 1:11pm

Wayne,

My understanding of the current rulemaking process (source: Patrick Egan on the ATTF podcast) is that they're thinking of a lightly regulated class (recreational, no COA, under 400 ft, LOS) of under 4 lbs. Pretty much like the current recreational exemption, but with a weight limit.

And yes, giving people better choices than circa-2000 hardware and circa-2000 prices (ie UNAV) was a big part of our motivation with ArduPilot. The UNAV guys have skills, so I hope they'll respond to the challenge with an updated approach to technology and customer service. If not, well, competition will do what competition does ;-)

Moderator Comment by Howard Gordon on April 2, 2009 at 2:12pm

Interesting board, but I suspect their data sheet is significantly in error about it weighing 40gm.

Moderator Comment by Jack Crossfire on April 2, 2009 at 2:33pm

10 years ago, single board computers with x86 processors were the way everything was done. In 1999 everyone was promoting web servers that fit in a deck of cards. Neural Robotics originally flew old PC104 boards. Unfortunately, it's the microcontroller which has replaced the single board computer since then.

Moderator Comment by Howard Gordon on April 2, 2009 at 2:43pm

It's been longer than that. My first full-sized robot and video tracking systems circa 1989 were built around 20MHz 80286 motherboards from Taiwan. Cost was around $100, which was a lot cheaper than PC/104. Compact PCI came out somewhere in that time frame as well, but they were also expensive. When these boards drop down into the $100-$150 range, they'll probably be broadly adopted.

Comment

• ‹ Previous
• 1
• 2
• Next ›
• Page