Micropilot Autopilot Install

[ErichF]

We are just about ready for the first test flight of our first UAV conversion of a Bergen Observer EB. We used the Micropilot MP2128heli autopilot with electronic compass, Ublox GPS, and Sonic Altimeter. Connection is by Microhard MHX910A 900mhz radio modem.

The system is pretty versatile, but not incredibly user-friendly to setup. Physical installation was pretty straight forward, though.

This system will allow completely autonomous take off, waypoint navigation, position holding, and landings. It also has built in support of camera gimbals and controls.

All for a modest $9000.

Enjoy the pics.

http://www.helifreak.com/attachment.php?attachmentid=31929&d=1196473828

http://www.helifreak.com/attachment.php?attachmentid=31930&d=1196473828

http://www.helifreak.com/attachment.php?attachmentid=31931&d=1196473828

http://www.helifreak.com/attachment.php?attachmentid=31932&d=1196473828

http://www.helifreak.com/attachment.php?attachmentid=31933&d=1196473934

http://www.helifreak.com/attachment.php?attachmentid=31934&d=1196473934

http://www.helifreak.com/attachment.php?attachmentid=31935&d=1196473934

http://www.helifreak.com/attachment.php?attachmentid=31936&d=1196473934

[WayneBrown]

looks like someone needs to get out the fab gear and make a G10 box to replace that heavy assed looking tin one.
other than that, pretty cool dude

[ErichF]

It's supposed to be made of metal, for shielding purposes. If I made a carbon one, that would have worked, also...maybe for the next one. Definately look cooler.

Made of plastic, I would have had to cover the thing in foil tape in order to shield it. Now THAT would have looked pretty hoky :)

[cbergen]

What's the all up weight, RTF?

[ErichF]

17 pounds with 1/2 fuel and a Sony Handi-Cam on the front, plus lead ballast for balance.

[cbergen]

Only 17 Lbs!! That's Excellent!! :)

[ErichF]

Yeah well, hold the mayo...

Tried a better scale today, as I was beginning to think 17 pounds seemed a bit light...

Uh, yeah, try 22 pounds. :shock:

[Crjbenny]

Erich,
Have you flow with the auto pilot yet?

[ErichF]

kinda sorta...we had some issues with GPS, datalink, and vibration of the sensors we have to work out. I'm in the process of re-doing the install with silicone gel-mounts and different mounts for the GPS antenna.

In process...

[rbort]

I was the test pilot for MIT for several years and then it transitioned into nascent tech for building and flying autonomous helicopters. MIT got the helicopter to fly aerobatic maneuvers, and later Nascent modified it to take off, fly to waypoints, return home and land, and put a whole bunch of other neat features into it.

Their system cost alot more than 9k but is very impressive in holding the heli in place and sending it places. I especially love the return home and land feature where you can command it or it can do it on its own if is loses contact with the ground station. Always impressive to see that come back like that, land within 10 feet of the takeoff spot and shut down by itself.

In any case keep us up to date on how this system works and you might want to consider putting an on-board generator on there to power up some or all of the on-board systems depending on the load you have. That way your heli can fly as long as there is fuel in the tanks.

-=>Raja.

[furyphoto]

Hi Erich,

I (and I think others) would love an update on how this system is working for you.

[ErichF]

We are waiting for parts from Bergen RC to convert the Observer to electric. Once that is complete, we will again start work with the Micropilot.

The only problem keeping us from flying at this point, is the fact we can't start the engine without causing a fatal error in the autopilot. The reason being the action of pull-starting the engine tends to jar the aircraft enough to cause the system to flag an error in the accelerometers, despite having the autopilot mounted on soft silicone gel mounts. Such a fatal error requires a system re-boot. We could get around this by using an electric top-start setup, I think, but we decided to try the electric conversion instead.

On the other hand, we have had great success with the same autopilot in a fixed wing Sig Rascal 110 ARF that we converted to electric. The integration was very straight forward, with the aid of the included hardware setup wizard. We had to make no gain adjustments from factory settings to get the airplane to autonomously fly box patterns, figure eights, and orbits in either direction. The electric conversion flew well for nearly an hour on two 10S 5000 Thunder Power packs in parallel. I hope for equal success with the E-Observer conversion.

[cbergen]

Parts are on the way. I had to remake the motor mount for the different gearing for 10s, and it may require a couple more gearing changes to optimize the setup, but should be cool nonetheless to prove the concept.

[ErichF]

Awesome, look forward to testing this conversion, especially if it flies autonomously. :smokin:

[furyphoto]

Thanks for the update, please keep us posted.

Also, if you get a chance, would you PM me, and let me know who you dealt with at micropilot, and how they were to deal with?

Thanks,
-A

[steven]

Hi everybody I unfortunately buy a MP2128heli in July 2007 and I received on November 2007, I think It is the first one they sell.
Are past 8 month and we can’t fly at all, we have got the same problem Erich has, we have complete resolve the vibration issue after many month of test, night at the phone with micropilot for gps issue and a lot of other things,:arggg:
Is like a infinity story.
There is anyone have experience with these system?
Erich have you got success whit these system, if you remember us why write you in private.
I ask to micropilot I Oder client can fly with these system, and they told me, one client can fly with a Berghen Observer are you?
Sorry for my English.
By Steven

[ErichF]

We are having great success flying the MP2128 in a fixed wing. It does a very good job of flying the airplane. Yes, we are the ones trying to fly a Bergen Observer with the system. Our installation on an electric converted Observer has passed RC hovering tests well, but just as we were going to turn on the MP2128 for Arcade Attitude Mode, the ESC went out on us. After two months, Castle is finally sending us a new 110HV. We will program it with a Castle Link as they suggested, and try it once more. I think it will fly the Electric Observer fine once we get it back in the air.

The GPS antenna they provide is very weak, and needs a ground plane that it can magnetically attach to, and is at least 3.5 inches square. Once we did this, GPS is great. Make sure to also remove the sticker under the antenna so the magnets can attach directly to your groundplane. We had to do this for the helicopter AND the airplane, it's not a matter of being under carbon rotor blades, like some have thought.

Good luck with you system. Be sure to update the software to the latest Beta version on their website. They have made some changes based on our feedback. The new Helicopter Setup Wizard works pretty good to get your helicopter base setup easy.

[ntustudent]

I belong to a group of students working on a project regarding UAVs for our University (Nanyang Technological University, Singapore). We came across the products Kestrel autopilot,piccolo and micropilot as candidates for our project to implement autopiloting in our UAV The UAV we are developing is not a fixed wing type, but a coaxial-rotor type helicopter.

What we need to know is that are there facilities to modify the autopilot flight control algorithm as we may want to implement some robust algorithm instead of the PID in the autopilot in these controllers.

pls help

[aghheli]

Hi,

I also have bought MP2128Heli in July 2007. I spent 5 months trying to manage vibration issue and finally I obtained 4 times vibration reduction. I use Hirobo Eagle 2 Long Tail helicopter as vehicle to my UAV project. This model helicopter uses 24 cc fuel engine which gives power of 2.8 HP. So, yes vibration issue was a big big problem. 'ErichF' do not be offended - this is not my purpose to offend you - but when I am looking at your frame and the way you mounted all Micropilot stuff, something tells me that it is not going to work (because of vibration issue). What is more I hope that if you replace fuel engin to electrical one your frame will be able to take off with all this payload. But true, electrical engines produces less vibration that fuel one. In my case I had to design tripple vibration isolation system which was preceded by forces and accelerations identification research. The engine I use has manual starter and I have to pull it a couple of time to get engine running. 'steven' right now I can not say: "unfortunately I bought this autopilot", but yes, this is true that this autopilot requires a lot of work which have to be done before any results are obtained. I have tested my frame in PIC mode a couple of times and I think that all parts have to be isolated from the frame (I use 24cc fuel engine which influences frame significantly). Last week I was trying to fly in CIC Attitude Arcade mode but something is wrong with rudder. Each time I took off my helicopter started to rotate about Z axis very fast. I am trying to resolve this problem.

Regards

[steven]

Hi aghheli you have to make a very very good compass calibration, the is the reason of your problem.
Make attention nothing metal is near the compass, is very sensible, if you want make a test, with Horizon Open and the system ready to star try go near with a little piece of metal like knife or a screwdriver, and see the helicopter moves around in horizon.
Have you eliminate the magnetic under the gps antenna?

[ErichF]

It could be the magnetometer calibration, but it's most likely the servo direction is reversed. Try simply going into the Servo tab of the VRS editor and check Swap next to the rudder servo. You will have to readjust your centering and travel range.

When we first flew our airplane with the system, we had the ailerons inadvertently backwards in the autopilot. The RC transmitter was masking the issue, as we had the channel set in the transmitter to operate the ailerons correctly, but as soon as we turned on the autopilot, it rolled right over. I suspect you have the same issue with your rudder.

[ErichF]

Better yet, download the latest Beta version of the firmware and GCS software. The new Helicopter Configuration Wizard makes setting up a helicopter very easy, especially when it comes to the throttle and collective lookup tables. Instead of the tables and raw servo values, the wizard uses curve points and percentages, a lot like your average RC transmitter. The Wizard has a checkout at the end to make sure all the servos operate in the proper direction, irregardless of the transmitter.

[aghheli]

Hi,

Both compass calibration and servo direction wasn't a issue which caused the yaw rotation. My GPS antena doesn't have any magnet at the botton.
I have investigated yaw rotation issue and I think I know what is the problem, but firstly I am going to verify this thesis.

Regards,

[ErichF]

Could you send me a copy of your VRS file so I can compare it with mine? I'm working with PID gains from MP that they used on their ECO-8, a much smaller heli. I want to see where your gains end up, as your heli is more in line with ours.

Cheers,

[aghheli]

Hi,

I have not changed PID settings yet so they are in configuration for Micropilot's ECO helicopter.
Have you replaced you fuel engine with electrical one?

Regards