Entering UAV's was never a preference of mine, since i'm more of a land lover. In fact, as a personal project I have an autonomous robot based on an RC Kyosho. Took everything off but the chassis, servo and the wheels. Recently, this guy that dedicates some time and money to RC planes gave me an idea... How about placing instrumentation on one of those planes so that I can measure the flight as accurately as possible.

The idea came to me because there was a spy cam attatched to the plane, and the image was very good, but for itself it tells me little information of the flight itself. So, I need to get some fight data.

After getting that information, the idea is to superimpose the data on the video, so the viewer can have a better notion on what was actually hapenning.

I prefer to start on an RC based platform for now, and develop the instrumentation to be acurate and liable. Something sturdy so that it can endure some crashes and live to measure another day.

I'm currently surfing the web for existing solutions and more knowledge. Any help, tips, pointers will be greatly apreciated.

To clarify things further, what I want to build is similar to a black box. Maybe later I can accept the chalenge of an autonomous robot, staring from the ground up.

Views: 8

Reply to This

Replies to This Discussion

You're just looking for telemetry data superimposed on the video stream? There are loads of commercial products that do that (EagleTree, etc). Just search for "OSD" (on-screen display)
Thanks Chris, but what i'm really interested in doing is building my own solution. It's not to be cost effective or anything, it's just so that I learn along the way, and have the satisfaction of building my own stuff. Plus, the learning process helps me professionally. I've had a class on microcontrollers, which didn't went as good as it could. The uC studied was the 8051C, but now i'm looking to work on PIC uC.

It's easier to go and get stuff from the shop, but that's not my goal. Right now i've come up with some possible solutions, some even came from this forum (it's nice to find help somewhere), and I will have to look into them. Still, commercial products always present a solution that sometimes is worth a look.

Once I get more stuff ready, i'll show my progress and post my doubts. I will also be on the lookout for doubts of other members. My still young skills might be of some help yet.

UAV's pose a dificult task for me, and I do have a question for you. How do you get the planes to fly fast and on course? My whole experience in robotics have been on studying ground vehicles. Reading the information of what's ahead to avoid obstacles and processing that information is already a chalenge in slow speeds like 0,2 meters per second. It takes fast processors and good coding. Is your aproach different from what i'm saying here? Or do you just fly the planes in obstacle free spaces? I must say it intrigues me a bit.

And i'm saying this because a recent friend has offered me the possibility to make an UAV on his chopper. Who knows? I might even join the club later on
We're more of an Atmel AVR community here, but for OSD you might want to look at the Parallax Propeller, which is easy to output video with.

As for avoiding obstacles, we don't ;-) That's the nice thing about being in the air. Between 400ft and 1000ft there isn't much there. We also choose GPS paths that don't run into mountains ;-)

Most of us don't do look-and-avoid at all...

PS--helis are HARD. You might want to start with fixed wing..
The nature of the uC for me is a bit irrelevant, I don't care much about that. For now my problem is to figure out which sensors are more suited for my needs, get or design the schematics and build them. Those sensors will work on whatever uC I choose, as long as I´m capable of programming it.

As for not avoiding obstacles, that was what I suspect. It involves a lot of expertise in different fields. That is why for now i'm only interested in telemetry. The reason is simple, these sensors (working) and the knowledge gained can be used in more bold projects. For what I intent to do, it would mean to develop something like the navigation system for the tomahawk missile. It's so ambitious that I classify it as impossible for the available time, resources and knowledge I posess. Hence the reason for dedicating myself more to land vehicles, where the speed can be tunned down to zero if needed. A true flying robot would need processors so fast that I would probably have to sell my house to buy one or two :)

The sensors I will need are similar to what you guys use. I would post on a robotics forum, but this one seems to be more complete than the rest I've seen so far. So, if having someone here that isn't quite on the same wave lenght, and still willing to learn and share with the others, isn't too odd for the rest of you, I can stay for some time.

I just thought that I should clear my objectives so that someone that doesn't feel at ease with it can make his or her point.

Oh, and I know about the extra difficulty of flying helis. It was an offer I received, and I immediatly talked to my friend to explain him precisely what kind of a challenge it would pose. I thanked him, but it's just too much for me right now.

Thanks a lot for the replies Chris. I hope that I can make myself useful to you or any other user.

RSS

Contests

Season Two of the Trust Time Trial (T3) Contest has now begun. The third round was a reliablilty/aerial photography round for both planes and copters, which is now closed. Stay tuned for the next round, beginning soon.

A list of all T3 contests is here

 

Groups

© 2012   Created by Chris Anderson.

Badges  |  Report an Issue  |  Terms of Service