The AttoPilot project

[Irv]

In early March I got involved in a "proof-of-concept" project that was to use an RC airframe and an autopilot system.
Goal of the first phase:

1) Demonstrate the ability of an RC airframe under autopilot control to perform tasks such as fly a specified path at a set altitude and airspeed, and initiate an event at a designated point along the flight path.
2) Learn as much as possible about the "care and feeding" and the potential uses of the AUV system.
3) Do all of this at as low an initial cost as possible.

After reviewing some material obtained by typing "UAV" on Google, I spent most of the effort in "RC Groups". I read almost all of "Beta Details - dmgoedde", "ArduPilot 2.0 software released --", "Picopilot", etc. I looked at many of the websites listed in "Sticky: UAV Links" and came to the conclusion that there are 3 main groups of autopilot systems suitable for use in RC airframes.

1) The low cost - labor intensive "open system" types.
2) The very expensive "commercial" types
3) The "under $1000 ARF" type.
I chose from door # 3 - the AttoPilot 8.1 by Dean Goedde.

Limited funds meant we had to use available airframes, computer, and RC systems. I have an Easy Glider that I used for FPV, a Futaba 9C RC system on 72Mhz, a video system with both 900Mhz and 2.4Ghz transmitter/receiver systems. Another team member donated the use of an HP laptop.

[Irv]

By mid March we were ready to go.
The Easy Glider had been modified for the FPV camera by placing a rear facing brushless motor on a pylon, putting the elevator and rudder servos near the tail, and carving out foam were needed for the video transmitter and the relocated RC receiver. BTW - my inspiration for the Easy Glider modifications came from "http://www.rcdon.com/html/the_sail-cam_project.html". Lots of good information there.

We decided to use the Easy Glider as a "flying breadboard". Implementation of the systems would be practical, not cosmetic. We sure met that goal - it's one ugly aircraft!

[IMG]http://www.rcgroups.com/forums/showatt.php?attachmentid=2543510&stc=1

[Irv]

The RC receiver is strapped to one side of the pylon and the AttoPilot strapped to the other. A tower was added for the XY sensor. The GPS antenna was stuck to the fuse at mid-tail, the MicroCard holder and Z sensor were stuck on the left side of the cockpit. The ESC was stuck on with Velcro.

[IMG]http://www.rcgroups.com/forums/showatt.php?attachmentid=2543713&stc=1

The system was interconnected as described in the "manual". None of the leads were cut. All excess wires were bundled and slip tied. We used a lot of double sided rubber tape, Velcro, blue masking tape and some CA glue.

[Irv]

The AttoPilot as received is a complete system. All sensors, all wiring, all software, and almost all the documentation are included. The documentation and software are provided on the MicroCard that will be used for flight control and data logging.

The CGS (Ground Control System) is outstanding! Much better than I thought it would be. It is used for after flight analysis at first. It will be used for realtime monitoring and control in the future.

http://www.rcgroups.com/forums/showatt.php?attachmentid=2543722&stc=1

[Irv]

The integration of the system with Goggle Earth was another very pleasant surprise. A plot of the flying area can be used to create the WP (way point) file and after the flight, the GCS can convert the LOG file from the flight to a .kml file that is used to plot the flight on the same Google Earth presentation.

http://www.rcgroups.com/forums/showatt.php?attachmentid=2543725&stc=1

[Irv]

HOW STEEP THE LEARNING CURVE or WHAT ATTOPILOT TOUGHT ME.

Read the documentation over, then read it again.
Be very careful when editing the SET and WP files. Minor errors can cause major problems.

I bought extra MicroCards. CARD1 is used for the initial "Test Mode", CARD2 is the flight card and CARD3 is a backup flight card. The new MicroCards need to be formatted and labeled. The card supplied with the system was preserved and protected in a vault. Not really, but I did put two complete copies of it on the home computer. One as backup and the other as the working copy for access to the software. That card is stashed is a safe place. Study the files from the supplied card very carefully. You will need the information contained in each one of them at some time.

The tuning process is very iterative. It involves adjusting some of the parameters in the SET file and possibly the adjustment or relocation of some sensors. Make sure the pitot tubes are properly installed and clear. It will take several flights just to get in the ballpark. Don't try too many changes at one time. Electrical noise can be a problem. I had to move the XY sensor away from the ESC and Motor wires. I also used a toroid around the ESC wires to the AttoPilot.

[Irv]

The "flying Breadboard" is to be used to evaluate all of the components of the UAV system before transferring to the next airframe. We kept a record of each flight and the changes made between flights. I made up a check list and mission log form.


UAV MISSION CHECK LIST

FLIGHT _____


DATE: ______ AIRCRAFT: _______________ LOCATION: _____________

TEMP:______ WIND SPEED/DIRECTION _____/_____ CLOUD COVER_______


AIRCRAFT PRE FLIGHT:

[ ] Control Frequency/ clear (pin) _____________

[ ] Check and record flight battery voltage. _____________

[ ] Check and record other batteries: _____________

[ ] Range Check

[ ] Verify proper CG

[ ] AttoPilot SD Card # installed _____________

[ ] RC Transmitter set to proper aircraft

[ ] AttoPilot mode selection verified: RC, ARC, LARC, AUTO

[ ] Verify number of GPS satellites ____________

[ ] Verify pitot and static tubes are clear

[ ] AttoPilot mode set to RC

[ ] Check aircraft control surfaces move are in correct direction,
deflection

[ ] AttoPilot mode to AUTO, Check control surface movement is correct
for Roll and Pitch

[ ] AttoPilot mode back to RC

[ ] Take off and start timer

[ ] Post flight. Record flight battery voltage. _____________






MISSION LOG

Aircraft Configuration: ___________________________
______________________________________________

Aircraft changes since last flight:
______________________________________________
______________________________________________
______________________________________________

Software (SET) changes since last flight:
______________________________________________
______________________________________________
______________________________________________
______________________________________________

Payload: _______________________________________
_______________________________________________
_______________________________________________

Mission Objectives: ______________________________
_______________________________________________
_______________________________________________
_______________________________________________
_______________________________________________
_______________________________________________

WP File configuration:
_______________________________________________
_______________________________________________
_______________________________________________
_______________________________________________
_______________________________________________
_______________________________________________

Flight Data: _____________________________________
_______________________________________________
________________________________________________
________________________________________________
________________________________________________
________________________________________________

Additional Notes: _________________________________
________________________________________________
________________________________________________
________________________________________________
________________________________________________
________________________________________________
________________________________________________

SET File logged as: _____________________________________

WP File logged as ______________________________________

LOG file logged as: _____________________________________

form updated 05/17/09

You don't need to have a laptop at the field but it makes the tuning process go much faster.

Make sure there is no LOG file on the MicroCard for the first flight of the day. After each flight you should rename the LOG file. If you don't the second flight will have its data appended to the same LOG file as the first flight. That is OK but that can make the Google Earth plot very busy.

I download the SET, WP, and Log files to the main computer after each session and save them as SET.date, LOG.date and WP.date
(ex. SET.05-24-09) That way I can recreate any session if questions come up later.

All of the first phase goals have been met.
1) Demonstrate the ability of an RC airframe under autopilot control to perform tasks such as fly a specified path at a set altitude and airspeed, and release a streamer at a designated point along the flight path.
2) Learn as much as possible about the "care and feeding" and the potential uses of the AUV system.
3) Do all of this at as low an initial cost as possible.

It has taken us 30 flight to get there.

AttoPilot has done everything we have asked of it!

Next - add the 9XTend TX/RX modems to the system for real time flight display and control
Add the video system and test for compatibility.

[Gary Mortimer]

Oh like the check list

[ios]

Magnificent !! Fantastic level of detail, and I too really like the checklist.

I'm really looking forward to more photos and videos, and details of your progress.

[Tom Harper]

Irv,

Great stuff, thanks!

I need the check list but never got around to doing one. I'll use it.

Tom

[airmcn_3]

Nice work!

[Gary Mortimer]

I have taken the liberty of copying and pasting Irvs checklist into a .PDF available over at the other place chaps.

Hope that does not grate too much Irv.

G

[KarbonBird]

Very interesting info - thanks for sharing.

BTW - what motor/ESC did you fit to the Easy Glider?

[Irv]

I used the Maxx Products HC2815-2000 brushless motor and a Castle Creations Phoenix 25 ESC and a 6X4 prop.

[aero_k]

Very interesting. So how much did you end up spending to get this far?

[Irv]

Very interesting. So how much did you end up spending to get this far?

I would have to estimate the cost because the plane and radio are a bit old.
Approximate costs:
Airframe - Easy Glider 100.00
Futaba 9c radio 300.00
Motor, ESC, batteries 250.00
AttoPilot 800.00
Misc. Chargers, etc 150.00
Laptop computer loaned
--------
1600.00

If you are starting from scratch - I would budget $2000.00 without computers, RF modem and airborne video system.

[airmcn_3]

I would have to estimate the cost because the plane and radio are a bit old.
Approximate costs:
Airframe - Easy Glider 100.00
Futaba 9c radio 300.00
Motor, ESC, batteries 250.00
AttoPilot 800.00
Misc. Chargers, etc 150.00
Laptop computer loaned
--------
1600.00

If you are starting from scratch - I would budget $2000.00 without computers, RF modem and airborne video system.


All your missing is the video system and modem for telemetry.... add another 2k if you want to do this.

[Irv]

We just had a very successful flight of the Easy Glider UAV breadboard.
In gusty winds and a fully overcast sky, AttoPilot held the flight path of 3 circuits of 7 way points very well and flew the circuit at the 3 altitudes, 60 Meters, 40 Meters and 20 Meters. The 20 meter flight path was very constant, much better than Dean suggested it would be.

All that and with a 2.4Ghz video system sending down great pictures. (Sorry - the DVR didn't store the video correctly - no video to share - yet.

Does any one (Chris) have any advice on the initial SET file parameters for a 12 pound, 9 foot wingspan Cub? We will be transferring the AttoPilot from the Easy Glider to the Cub soon.

[airmcn_3]

We just had a very successful flight of the Easy Glider UAV breadboard.
In gusty winds and a fully overcast sky, AttoPilot held the flight path of 3 circuits of 7 way points very well and flew the circuit at the 3 altitudes, 60 Meters, 40 Meters and 20 Meters. The 20 meter flight path was very constant, much better than Dean suggested it would be.

All that and with a 2.4Ghz video system sending down great pictures. (Sorry - the DVR didn't store the video correctly - no video to share - yet.

Does any one (Chris) have any advice on the initial SET file parameters for a 12 pound, 9 foot wingspan Cub? We will be transferring the AttoPilot from the Easy Glider to the Cub soon.


Irv,

A couple of things you need to do with a larger bird is increase the steer and alt prop band width.

Your gain settings will probably need to be increased due to how slow a cub reacts to aileron input.

Get your expo settings above 25

That’s about all until I see a SET and LOG file.

Thanks,

Chris

[jginsandiego]

Irv/ anyone?
just got registerd on this site and had a question I also live in Southern Ca and was wondering if you in your trials have seen any issues with the thermal IR xy control? I see more then just ATTO useing it for the XY control.

I ask because it would seem that if one sensor looks out at the ocean for instance and the other is say looking at a mountain or anything that is of a different temp I would wonder if that could confuse the sensors and cause control issues, living by the ocean we get fog and some rather interesting weather and if I get one of these I dont want any problems.

same question but lets say were in the Dezert and its pretty hot out over a 100 F. do think that could cause a issue. I cant find the specs on those sensor units same look as this one from another system

http://www.fmadirect.com/detail.htm?item=1489&section=20

The only other autopiliot system I see that is using a different type xy sensor is picopilot they use a barometric pressure sensor.

just trying to learn the differance and what good or bad things have been seen with both types.

[jginsandiego]

Below found this to answer some of my question and anybody who may wonder the same but would still like to here what others have to say about theroy or real life use of a thermopile XYZ stablization system.

http://paparazzi.enac.fr/wiki/Theory_of_Operation

[dmgoedde]

I live near Phoenix in Arizona, and purposefully designed the XY and Z themropile heads to not saturate their output even with 130+ F ground temp. Also has been tested to below 32 F (0C).

Water on one side of the aircraft is not a problem. This has been thoroughly tested. Atto uses GPS and barometric data to constantly trim a pitch and roll bias. This is why mountains and rough terrain don't matter either. This is all being said based on lots of real test flight by myself and others.

All that being said, thermopiles can't compete with an IMU for low latency angular rate and attitude data. You really wouldn't notice attitude latency in thermopiles unless you have flown both them and IMU. There is a subtle but distinct performance advantage for the IMU.

[jginsandiego]

Thanks for the reply is there any way to get a copy of the instructions that come with a Atto unit my friend JOHN S. went to go see Irv fly his unit last Friday and we are both looking to get one. I want to build my plane I will use first and I have to save up for the Atto would like to read up on the setup and placement of the sensors and tech info I did not see anything on the Atto web site for instructions.

[airmcn_3]

Needs to be updated.....


Chris

[dmgoedde]

Bill Nailor (whom I mentioned earlier) is working for me to polish the manual. He has given it a lot of time and it is very much scrubbed up now. This will be published soon, as within a week or two.

[jginsandiego]

Thanks I will look forward to reviewing this where will the updated one be when it is done.

[Gary Mortimer]

Here I hope http://www.attopilot.ning.com