Moderator Comment by Morli on December 31, 2009 at 4:14pm

Tj, That landing was as good as it gets in my book for autolanding sequence :) Landing was at constant ange of approach , impressive, how does it behave with mild wind in autoland mode. Many can't do that in manual mode :)) . good work

Comment by UFO-MAN on December 31, 2009 at 4:34pm

Nice.

Do you use a pitot tube and pressure sensor for avoiding stalling on landing? (holding the pitch of the plane with the autopilot while slowing down too much will eventually stall it).

Do you use GPS for height indication?

Do you have other height sensors, like e.g. a barometric sensor?

Can you give some more details of your landing algorithm?

UFO_MAN

Moderator Comment by Morli on December 31, 2009 at 4:46pm

How about using a ultrasonic sensor comes in play for last 10' , why is this device not used in any if DIY AP? Can any one pls enlighten me pls regarding issues of using ultrasonic altitude sensor for first /last 5 mts ? How to use it in current boards in some manner, any idea is appreciated. Thanks. .

Moderator Comment by Morli on December 31, 2009 at 4:58pm

TJ, How did you dwploy the flaps? what channel , which code? Would readlly help me to find a way to use the flaps( I am flaps fan) , thanks

Comment by william emery on December 31, 2009 at 6:43pm

Really good, and I like your approch to the problems of a uav's. My problem is that I'm about as new as you can get and I read all the informatiom you put on your web, but Im still in the dark. Are you going to make all this information available to share, sale??? anything???

Moderator Comment by Morli on December 31, 2009 at 7:40pm

Hi William , welcome
If I may say so , most if not all info here is free and effort is DIY :) , kits and parts ofcourse are for sale from number of kit makers :)

Comment by Tj Bordelon on January 1, 2010 at 7:25am

@Moril: Thanks! It's good, but it's still landing too fast with the old way. There's usually minor dings from the landings, which isn't good. This code has landed in 20 mph winds, and the high speed helps in those situations. The new landing code hasn't been tested in high winds... I'm eager to see what happens!

@UFO-MAN- I have a pitot and it "backs off" of the commanded pitch angle as the airspeed slows below a threshold. That is soon to be replaced with an airspeed command to elevator feedback loop. It's very cold in boston and I just added much needed calibration to that sensor. So-called calibrated pressure sensors from digi-key were off by as much as 10 mph when in cold weather! I use GPS for height, and I know the error at all times (DOP) so I add a fudge to the flare altitude.

This is the new code that loiters then lands. I think the ultrasonic sensor idea (@Morli) would let me be more agressive on the approach since I know exactly where the ground is. I could hold a negative AOA until I detected the ground, then go to a positive AOA for flare. But right now I just hold a positive AOA with flaps (which gives me a constant airspeed) until I finally land. But this makes the landing take a few hundred feet more space.

My stuff is custom (http://bordelon.net/) so there isn't really a channel on my setup... just a jack for alierons which I offset downward for flaps.

Also, a word on flaps-- Some prefer spoilerons (which are flaps that go up) because it avoids stalls and is more effective. I agree, but the downside is that the plane comes in at a very high AOA and the tail hits the ground. On the easystar this tends to cause damage!

Comment by Tj Bordelon on January 1, 2010 at 7:30am

@william - I don't know what I'm going to do with my stuff. There are plenty of others that seem to want to sell into this market. I would love to sell UAVs ready-to-go but my understanding is that you can't do this legally right now :(. The market for DIY is too small for my tastes right now. Maybe one day?

Comment by william emery on January 1, 2010 at 7:59am

I know electronics but can't program a light bulb, but I have a pic programmer, would the the UAVDeBoard be a good start???? I've played a little with picaxe, and pic in basic on simple robots. It seems that programming is the key to success.

Comment by UFO-MAN on January 1, 2010 at 9:02am

Hi Tj!

I think I understand your AOA "backoff algorithm" based on airspeed threshold.

I assume what you are going to do in the next version is to make a PID loop that controls the airspeed with the AOA. I guess you then will have certain max and min limits for the AOA that you will not exceed.

Would you be interested in sharing the Pitot tube design? I guess the pitot tube needs to be corrected with both air temperature (density) as well as static barometric pressure offset (high pressure vs. low pressure days) ?

You may want to take a look at the pressure altimeter discussion going on on my blog page as well. We discuss using Kalman filters for an implementation.

UFO_MAN

Comment

• ‹ Previous
• 1
• 2
• Next ›
• Page