Manned ground vehicle

[clolson]

Ok, manned ground vehicles aren't quite as exciting as unmanned aerial vehicles, but you have to crawl before you fly, right?

I just wanted to report on a couple small success I had this weekend.

1. I made some really good headway on my code to drive the Xbow MicroNav.

2. I added some functionality to output the gps, nav, imu, and servo (transmitter) data packets over a serial link with start packet markers and strong check summing.

3. I wrote a separte utility to record the serial output from the micronav, also decode it, and pass the position over to a live running "slaved" copy of FlightGear to draw a 3d synthetic view of whereever the micronav is right now.

4. I got a pair of maxstream radio modems configured and running at 115,000 baud. (Now life is getting good.) :-)

5. I installed the micronav and one end of the maxstream in my car.

6. I located the other end of the maxstream pair in my living room connected to my laptop running the decoding/recording software.

7. The decoding/recording software sends the data over to FlightGear running in my basement on my desktop PC.

8. FlightGear was connected in the FlightGear multiplayer system which means anyone, anywhere can see me via the FlightGear multiplayer google map system.

9. Long story short: On my PC downstairs I can watch a 3d synthetic view from the perspective of my car and anyone, any where with a web browser can track me in our google maps ... all live.

The only down side to this all is that I was driving so I can't see myself move on the map at home.

10. I'm going to leave the system running over night for testing purpose (hope I don't run out my car battery!) So you can observe my car sitting in my driveway (hopefully) ;-) by going to this link:

http://mpmap02.flightgear.org

Then click on the check box next to the Jetta1 call sign (97 Jetta in case you are interested.) Sometime tomorrow/monday morning (USA Central time) I'll have to shut this all down and return all the toys to work.

So there you have it ... my fun and excitement for the weekend. :-)

The next step is to put the Xbow MNAV and maxstream in an airplane (probably one of our Rascal 110's) and do some passive flying and data recording (the MNAV would be outside the critical control path, just riding along collecting data.) The goal is to gain some confidence in the MNAV in the air, wring out the maxstreams, get an idea of the quality of data we get back, etc.

If all that looks good, then we will be starting to dabble in actual autonomous flight before the year's end (I hope) :-)

Curt.

[lvspark]

Darn , Looks like I missed it.. Cool stuff Curt, keep up the good work!

[HELModels]

I checked it out and noticed the heading was spinning like a top. Actually the heading changed about one per second, sometimes held steady. It was indicating a change in computed position, move your position to the E or W, N or S, requires a heading change.

[Unterhausen]

I checked it out and noticed the heading was spinning like a top.
I had that happen before I calibrated the MNAV. The gyros are particularly difficult to calibrate correctly.

[clolson]

The observed heading was due to the FG multiplayer system. It computes your heading based on position - last_position not reported heading, so when you are stationary, but the GPS location migrates around a bit, your heading (in the FG multiplayer system) will change pretty randomly.

[treehog]

excellent stuff mayby You might consider to a build thread at a suitable forum and link back to this thread as the UAV and UAS have lots of similarities

I have opted to start a thread on a similar theme at the robotics section (sub forum thinking robots)

http://www.rcgroups.com/forums/showthread.php?t=577677


A build thread near there could be very useful if you can


Ralf

[clolson]

excellent stuff mayby You might consider to a build thread at a suitable forum and link back to this thread as the UAV and UAS have lots of similarities


I do have a build thread (and unfortunately a rebuild thread) here:

http://www.flightgear.org/~curt/Models/Special/Rascal110_1/

and ...

http://www.flightgear.org/~curt/Models/Special/Rascal110_2/

I haven't started anything specific for the flight controller yet, but I will. Basically I'm developing an open source software package to run on/with the xbow micronav hardware that will be much more sophisticated and flexible and usable compared to the default demo code they ship. I've got a really good start going, but I want to get a little further along before I make anything generally available.

Curt.

[clolson]

We have begun to make some progress on our Crossbow MNAV and Stargate integration and packaging along with a Maxstream radio modem. We are installing this in a Rascal 110 so we have a fair bit of space to work with. Here are a few pictures:

http://www.flightgear.org/~curt/Models/Special/MicroGear/

Right now we have some concerns that the Xbow provided attitude determination code is not all that great (or might have some subtle mistakes) so we are working on digging through that in parallel to everything else.

Regards,

Curt.