Cornell Builds Autonomous UAV

tshak writes "From Microsoft Research, 'Faculty and students at Cornell University have built an unmanned airplane with its own on-board, embedded control system. The large-scale model plane flies by accessing coordinates from an off-the-shelf GPS unit.' Not only does the plane run XP embedded, but the software is written in C# on the .NET Compact Framework. This is all powered by an 800mhz Crusoe processor with 1GB of total system storage."

1 GB?

[nullset]

"only" 1GB? that's hardly impressive. Try fitting that in a few kb of ROM, like a real product would be, and i'd be impressed....

seriously, this is nothing but a joke. avionics software does NOT use 1GB of storage, nor does it run a PC-oriented OS (be it XP or linux).

This is quite silly from an embedded point of view. I really hope they weren't hoping to impress embedded developers with those specs.

--buddy

Instead of the usual...

[Anonymous Coward]

Instead of the usual set of jokes about blue screens of death from the blue skies, etc., can we get someone who knows about embedded XP systems to comment? How robust can this system be? Also, given that you can limit the number of inputs (e.g., the plane won't be browsing the web), can the limited number of apps it has to run help with stability, which tends to be an issue in desktop XP with many apps open?

Wow

[kalirion]

11 out of 14 posts so far have been modded down as Flamebait, Troll, Redundant, or Off Topic. Is this usual? In any case, how long until these little plains are fitted with a collision detection and avoidance system? Flying based on checkpoints is all fine and well if you're the only one in the sky, but othrewise its equivalent to a blind person without a dog or a cane walking down a street he knows very well.

Microsoft research aides terrorist organizations

[Anonymous Coward]

all off the shelf stuff huh ? now imagine if this device carried a payload

Trans Atlantic Model

[naznerd]

Small GPS controlled planes are old news. Check out the TAM project. Trans Atlantic Model. They flew a model plane across the Atlantic Ocean last year. Check it out here -> http://news.bbc.co.uk/2/hi/europe/3145577.stm [bbc.co.uk]

Re:Kick back?

[timmyf2371]

Nothing so "shady"...

Award of academic grant. [microsoft.com]

The article itsself states: Last year, the group won an Innovation Excellence Award from Microsoft Research to continue their previous work in designing an autopilot system for a large scale model aircraft. Schools around the globe received awards from the Microsoft Research University Relations program to enable them to conduct research in emerging technologies.

So they are up front about it - I personally think this is interesting - doesn't matter that they used XP-Embedded over another operating system - whatever gets the job done.

Re:Kick back?

[CausticPuppy]

Why do most people here seem to concentrate not on the fact that something very cool was accomplished, but rather on the fact that it was accomplished using technology from a vendor they don't like?
</rhetorical>

How about commenting on the project itself rather than posing conspiracy theories? The article itself even mentions that the team had gotten previous awards from Microsoft.

Re:1 GB?

[mr_z_beeblebrox]

Embedded software may not typically use 1 GB of software. But I would not find it comforting to think that the full instruction set for an airplane was able to fit into 640 KB of ram.

Re:Cruise missile

[Rosco P. Coltrane]

And the difference between this and a cruise missile is what exactly?

Cruise missiles cost billions in taxpayers money.

Mean-spirited

[Snart Barfunz]

Not to come over all gee-whiz and so on, but how relentlessly negative these posts are. The students deserve some congratulations for successfully completing an impressive piece of work. Maybe they didn't go the most efficient/difficult/brag-worthy route. So what? Everything they've learned will be useful, regardless of what hardware/software they end up using in the 'real' world. They probably had fun and have achieved something real, instead of just sitting back criticising.

Good point about the cruise missile though...

Re:1 GB?

[Potatomasher]

A few KB of ROM ? Common we're not in the 80/90s anymore... Some guy on this page talked about overdesigning. But there is such a thing as "under-designing". A small micro, 32 MB of flash and RTAI Linux will run you for only a couple of dollars and make your life A LOT easier...

Re:Kick back?

[yoshi_mon]

Well for one thing this is Slashdot after all.

For another most of the posts here, correctly imo, point out that this is nothing new or even that cool. Other teams have allready accomplished more than this with much less overhead. Why do you need XP on a plane? To play minesweeper on?

Lastly I doubt many people here would consider an "award from Microsoft" as being much of anything to brag about.

Re:Cruise missile

[clintp]

And the difference between this and a cruise missile is what exactly?Cruise missles aren't designed to land and take off again another day.

Joke if you like....

[SilentReproach]

But XP embedded is actually a very useful product. It is something that should be released as an option to run desktop systems, as it can be modularized and stripped of nearly all Microsoftiness (Messenger, IE, you name it). Just want the XP OS with full GUI, no frills? XP embedded fits the bill. We use it for a custom application here and it's just what the DOJ ordered.

Re:Instead of the usual...

[TK2216UKG]

A bit like, say, Linux then. A shame that all Windows installations don't work this way.

Re:And I'm supposed to be impressed?

[Rosco P. Coltrane]

I'm sure they could start from the codebase they have now, work really hard, and have equivalent expertise built into a lighter package in some number of staff-years. Alternately, they could archive the source, go drink margaritas for a couple years, and then buy the lighter package with equivalent power off the shelf and use it to run the existing already-completed software.

Oh yeah, that's just what we need, engineering students whose mindset is that of a business owner. How about busting your butt to achieve something? researching more elegant solutions (and no, that's not yours, elegance is in the design and performance software with tough constraints)?

You totally missed the point of research. Research isn't about using technology, it's about inventing new technology. Using off-the-shell components doesn't push the envelope, it just shows the Cornell students can take envelopes from their sponsors.

Good thing not everybody thinks like you, otherwise we'd all be waiting for everybody else to solve our problems for us...

Re:Ugh.

[Anonymous Coward]

run .NET on a system that doesn't need it

In this case, a type-safe runtime environment like .NET (or the JVM) is well worth it. The added runtime overhead (in terms of storage and processing) is insignificant in comparison to the risks posed by a simple memory-access bug, such as a buffer overflow.

Re:Mean-spirited

[philbowman]

Yes - at least they produced something original. :-)

Congratulations to them.

OTOH, if they had used Linux, everyone would have just gone on about how great it was that they'd used Linux, not how clever that they'd got the UAV to work. Can't win...

Re:1 GB?

[mr_z_beeblebrox]

Engine control software decides what the engine will do based on what the operator is doing. The controls for a UAV ARE the operator. Big difference.

XP is pretty doggone stable...

[Anonymous Coward]

I am probably one of the biggest anti-MS Linux biggots on the face of the planet, but I also have to run Windows for my job, and I also run it at home to play UT2003/2004 because quite frankly the game is slower than next Christmas on Linux. I've got an overclocked P4 2.4 Northwood running at 3.2GHz on air cooling only, that's been up continuously 24x7 since mid-March 2004 running Folding@Home all the time, plus I play UT on it and just leave the FAH service running in the background since it basically goes to sleep when something else is running. I've had *zero* operating system crashes, maybe 3 or 4 Unreal game crashes that never bring down the o/s, and I play UT every day for at minimum a couple hours (and my hands are just about crippled now to prove that :-) ). FAH has never errored out either nor has it ever had to toss out a work unit either. There are no signs of memory leaks either. It's been rock solid. Of course I have not patched the O/S with Windows Update either since then, but it's immune to attack from the Internet due to sitting behind a nazi-like secure Linux iptables firewall too.
Now that's only about 5 months of continuous uptime, but you gotta admit, that's pretty good for a Windows desktop box.

Re:And I'm supposed to be impressed?

[Zaiff Urgulbunger]

But one could argue that the research and development in this project was the software. Therefore it perhaps makes sense to use existing, proven technology so get the proejct off the ground (sorry!). Later refinement would then be to reduce weight etc, so *then* it might make sense to develop lighter hardware.

Re:1 GB?

[dasmegabyte]

Oh of course -- because a technology demo -- like a UAV that flies itself-- is completely unimpressive unless it's hyperefficient and has a tiny footprint.

Give me a break, man. It's a STUDENT PROJECT. They probably had a few months to do it. And you expect them to build their own OS and heavily optimize code? The REASON they used embedded XP/C# (besides getting them free from MS) was that they are efficient environments for rapid development that are easy to learn. There's very little bizarre apocrypha in C#, and when you're strapped for time that's way more important than impressing embedded developers.

Is XPE suitable?

[serviscope_minor]

Interesting project, but can anyone comment on this?

A control system really needs to be hard real-time. Is XPE hard realtime? Also, is C#? If it is, how does it deal with garbage collection under these constraints?

Re:And I'm supposed to be impressed?

[dave420]

Have you even seen an 800mhz crusoe and 1gb of storage recently?? You can fit both in the palm of your hand. We're not in 1983 any more. Complaining about the size shows how hard you're clutching at straws. I mean, sheesh. If it was linux you'd be all over it. "oh it's so cool! i love it! linux rules!". Every single article on /. has some assclown calling something microsoft-based crap, for absolutely no good reason whatsoever. The most disappointing thing is the rest of the /. community doesn't correct such blatant assclownisms, but eggs them on. Really, really sad.

Have you read about XP embedded? It's a pick-and-choose OS, so you can select exactly what you want. That means no bloat. Absolutely none. Kinda destroys your ill-conceived argument, and shows it really was a rant against Microsoft.

Re:Kick back?

[Anonymous Coward]

what the hell does it REALLY have todo with "slashdot" get off the idea and setting the trend you retard. This is tech.. and i wanted to know about it...

"News for Nerds. Stuff that matters" If you want please have slashdot change the slogan to.

"News for linux. Only linux news matters."

Some of you are not making life of open vendor any easier then microsoft. you guys are just as bad.. just you claim to be the "li'l guy"

man, some of you linux geeks are like.. let me show you a picture .. shows picture of "li'l brudder" now do you know how he makes it in life.. he uses linux......

get off the excuse of THIS IS SLASHDOT. man, own up and admit you are a zealot..

Proof of Concept - Scary

[bill_mcgonigle]

"only" 1GB? that's hardly impressive. Try fitting that in a few kb of ROM, like a real product would be, and i'd be impressed....

I'll take my flight-control software with an exception handling system, thanks.

What's more interesting is this is a good proof of concept that anyone with a few thousand bucks and not even a college degree can put together an autonomous airplane with commodity parts.

I think people who can get C4 can come up with the money for a gig of RAM. This kind of design can scale pretty easily, therefore so can the lifting capacity.

If one of these was launched from Hoboken could it be shot down over the Hudson River? Are they even considering this problem? If all the vans and trucks are being stopped in the Holland Tunnel it doesn't make sense to use the Holland Tunnel.

Re:And I'm supposed to be impressed?

[sexylicious]

Oh yeah, that's just what we need, engineering students whose mindset is that of a business owner. How about busting your butt to achieve something? researching more elegant solutions (and no, that's not yours, elegance is in the design and performance software with tough constraints)?

You are forgetting that a good engineer is, by nature, lazy. The only reason we have cars, planes, trains, boats, and so on, is because an engineer was too lazy to walk, try and fly (by flapping his arms), too lazy to run, and too lazy to swim. Engineers typically say to themselves, "How can I do this with less effort/money/time/etc.?"

Re:And I'm supposed to be impressed?

[m00nun1t]

Yeah, I bet they used an off the shelf processor as well, rather than building their own from scratch. And the metal used in the frame? Don't tell me they BOUGHT that rather than mining it and developing their own metalurgical refining processes.

The problem doesn't always lie at the level that happens to be of interest to *you* - the OS in this case.

Comment removed

[account_deleted]

Comment removed based on user account deletion

Re:1 GB?

[HeyLaughingBoy]

Big difference.

Not really. Engine control software "decides what the engine will do" based on what the operator desires, not blindly following what the operator is doing. The operator isn't sending 1000 FIRE signals to sparkplugs every second; the operator may want to rev to 8000 RPM, but the ECM will cut fuel at 6000, etc.

The ECM translates operator requests and adjusts for programmed limitations, current sensor states and engine control parameters to try to fulfill those requests, the UAV controls are doing essentially the same thing, but the operator requests (go from point A to point B) have been programmed in before flight started.

Re:And I'm supposed to be impressed?

[poot_rootbeer]

Oh yeah, that's just what we need, engineering students whose mindset is that of a business owner.

Actually, we do. Real-world end results are what actually matters.

How about busting your butt to achieve something?

I've known a lot of engineering students at Cornell. Suggesting that they don't "bust their butts" or serve as puppets of the Microsoft PR machine makes me laugh. A lot.

Think about this -- if the Cornell UAV team hadn't accomplished something unprecedented, would it have been "news for nerds"? Frankly I don't see why you care whether they developed new technologies from scratch, or built upon existing technologies.

Re:And I'm supposed to be impressed?

[Fujamabob]

The plane was supposed to also do recon. You want to do image processing in a year as a side project as an undergrad on custom microcontrollers? You also want that microcontroller to coordinate communication between the plane and ground station? If it were just supposed to fly, they'd have engineered it appropriately.

Give the kids some credit.

Comment removed

[account_deleted]

Comment removed based on user account deletion

Re:And I'm supposed to be impressed?

[Blakey Rat]

And the parent is the very definition of "assclown." Holy shit, man, get a grip. It's a piece of software, not a religion.

Re:Did they read the eula?

[Grishnakh]

Is a model airplane crashing really that much different than a store being shut down for 24 hours because their computers are down?

Yes, because this plane is large enough that it could potentially hurt or kill someone if it crashes. What if it came down on a busy freeway, causing an accident? A store being shut down only causes financial problems for the store.

An application like this has no business using a poorly engineered, consumer grade operating system with a long history of faulty performance.

Re:Ugh.

[mingot]

Maybe, just maybe, to those people the time saved using off the shelf hardware and software is more important than being able to thump their chests and brag about how tight the assembly code they can write is? Now, I know that YOU can write bug free assembly in your sleep, in half the time it takes us mortals to do it, so you don't need to mention that in your followup.

Re:XP is pretty doggone stable...

[Anonymous Coward]

Just because it's behind a firewall, doesn't mean it can't get infected, etc. It really needs the patches to become secure.... just like linux boxes need to patch the kernel every once in a while and require a reboot.

Re:1 GB?

[dasmegabyte]

First: I am embarassed that you think the secret to tight, efficient code is the elimination of bloat. The whole point of bloat is that it isn't used. If something isn't used, it's not necessarily inefficient. In fact, if space is not an issue, it would be an inefficient use of time to eliminate unused components.

Besides, the point of a project like this is not to TEACH anything...it's so the students themselves can learn how to engineer a solution given a set of constraints. If the constraints are "build an autonomous UAV linking off the shelf parts in three months," and they have at their disposal a system which NATIVELY integrates all of the available parts without the need to write drivers or compile special libraries, it would be a disservice to force them to use something else. It's be like telling a carpenter to build a shed in three days and that he has to mill his own lumber. It's unfair to say "you can't use this program because it has a few functions you won't use, making it bloated and worthless."

Tell me: when you're hanging a picture, do you refuse to use a claw hammer just because you don't intend on pulling any nails out? I mean, the claw is just added bloat right? Makes the hammer heavier for functionality you probably won't need.

Fucking retard. There's more to life than systems that fit on a floppy diskette. This is why we INVENTED the computer as an abstraction to calculation, C as an abstraction to machine language, etc.

Re:1 GB?

[misterTreellama]

The problem with slashdot is there are far too many programmers with tunnel vision. I was in a UAV club in college, and the operating system was the least of our problems. It takes an enormous amount of time just to keep the aircraft flight-worthy (especially after botched landings), and fiddling around with the mechanics enough to fit your processor, GPS, radio modems, flight sensors, cameras, and 9 tons of NiMH batteries on the tiny-assed vehicle. Finally, since you're just college students, you spend absurd amounts of time begging for "test" parts that you never intend to give back to the manufacturers. Damn right they're going to use an off-the-shelf OS, especially if they didn't even have to beg.

Re:And I'm supposed to be impressed?

[Moofie]

How about finding the best available solution considering the time and budget available?

Good engineers deliver. Great engineers come in on time and under budget.

What you propose is fine for "It never really needs to work" academia, but not acceptable for engineers who actually have to produce functioning hardware.

Re:And I'm supposed to be impressed?

[tshak]

Keep in mind that the entire plan weight less than 13lbs, so the 800mhz crusoe was not weighing the thing down. Also, I think this was more of an excercise in software, not hardware.

Re:Kick back?

[neonv]

I'm a grad student researching UAV's just like the one that Cornell is using in this articles. There are only a few major differences between what we do and what they do.

1. We didn't use .NET or any other operating system that costs money. The plane has an embedded system that doesn't need much operating system power. Most of what is run on the processor consists of feedback loops that are run under time constraints that are not hard to meet. Image processing is the only intensive processing and shouldn't be done by the operating system if you're doing research on it, so there is no need for a costly embedded system on the plane. Our hardware consists of a 30MHz processor (on a custom board), 512 KB (not MB) of RAM, an RF modem, and a gps unit. All of this is on a light plane made by a few engineers at the university. Using that, we do everything Cornell does at a fraction of the cost and complexity.
   
2. Planes fly themselves by themselves all the time. It's an autopilot. We've had one for a while. It flies and lands by itself. It sends telemetry information back to the ground station so we can monitor it and/or render it in an open source flight simulator that we modified for our use. All the processing on getting from one place to another, however, is done by the plane itself. It does multi-agent coordination by itself as well.
   
3. We use our own device drivers. You'll find that in an embedded system, drivers are not nearly as hard to deal with as a PC. Drivers just aren't a problem. You know you're hardware too well to have problems with it. not only that, but there are very few devices to work with in comparison with a PC.
   
4. There are lots of other differences, but they're not important. I see Cornell doing is spending a lot of money on fancy gadgets and software that they don't need making their system more complicated.

However, I do like the fact that they do image processing on the plane. In the case that image processing is needing and a ground station is not available, having onboard processing is needed. But once again, this does not require nearly what they are paying for.

Re:Ugh.

[andreyw]

Actually small footprint does not imply hardcoding everything in assembly. Small footprint would very likely be a tiny forth virtual machine + bytecoded forth. If you can't grasp what I am talking about - think OpenFirmware/OpenBoot. Forth is portable, so moving everything to a new architecture would mean porting the (tiny) forth VM - that beats porting XP. And forth is a pretty damn powerful language... Hell, if you felt magnanimous - you could even add an interactive forth interpreter for debugging purposes, and none of this would require 512 MB of memory.

Remember, if this wasn't a plane, but a *space* UAV, then those 512MB wouldn't come cheap - considering they wouldn't be off-the-shelf DDR DIMMS, but something like space-radiation-hardened SRAM. And uh... no offense... but with Windows' track record... I would not be willing to shell out $$$ for something that some 5r1p7 1dd13 would have fun with for 5 minutes before causing a catastrophe.

"Look Ma! I found a new cool flight simulator! Whoa! So realistic! Is that a nuclear power station? Coool."

Re:Ugh.

[ThosLives]

Sorry you missed my point about "management direction" above. I'm not saying that I can write bug free assembly, or even that assembly is required (OSEK and many RTOSs are written in C). Basically my point is that it's a sad day when people say, "ah, the hardware will handle it, and I've got a toolbox which will let me be sloppy, so what the heck?"

sure, for fast things, being able to throw lots of money and buy hardware is nice - but that assumes you have lots of money (or, perhaps, a hard-/software vendor giving you free stuff to use for good press and to get people familiar with their products). That's no substitute, though, for well thought-out solutions and careful code. I'm not saying that these folks at Cornell didn't do good engineering; my comment was on the seemingly obscene amount of hardware they used. Like most things in the "modern world" our problems - even things like viruses - aren't technical so no amount of technology can solve them. Until folks realize that technology cannot solve social issues, people will continue to be disappointed when technology doesn't solve all the problems.