Thinking Of Finally Building A Uav

[mecevans]

i have been thinking of building a uav for a senior project for quite some time, and now that i have a job i have money to do it. my budget is about $1000 US and i need some idea's. i would like the plane to be able to fly to a point and take a picture and fly back to the launch site with realtime data to my laptop.

i have allot of experience in aerial photography and video with a slowstick i built a year ago. at one point i was very interested in the paparazzi autopilot but now i see that there are two other cheaper autopilots, the ardu and the atto. i have about 2 months to build,test,tune and complete it. any idea's would be appreciated, feel free to put me on the spot! :D

thanks,
michael

[Buzz_Lightyear]

As far as I know, the Atto is not nearly as low-cost as paparazzi. Even if you don't want to build it yourself. Correct me if I'm wrong, but this is the order of the most popular, by ascending price:

Ardu - Paparazzi - Atto

(note: this is just for the autopilot itself. you can get all kinds of expensive with link and video equipment)

[rich smith]

The Ardupilot is not operational yet and probably will never work as autopilot with GPS altitude alone.

Attopilot is fully tested and full featured. Expensive but within your budget. Maybe not available commercially yet?

Paparazzi is strictly for DIY masochists.

[rich smith]

I should add RCAP to the list. I believe RCAP3 does what you want and requires little effort compared to the other two DIY approaches. Not sure if the waypoint sequencer has been integrated yet or still a separate board.

I know more than one person has gotten a plane to do what you describe and it's a fairly mature project. Search is your friend.

The Ardupilot is not operational yet and probably will never work as autopilot with GPS altitude alone.

Attopilot is fully tested and full featured. Expensive but within your budget. Maybe not available commercially yet?

Paparazzi is strictly for DIY masochists.

[Tuner]

Paparazzi is NOT!!! strictly for DIY masochists, but Rich sure has a good point. Its a little daunting if you don't have help from the community.

If you get involved in the PPZ community you will see that while it is a complex autopilot and is not a 123 step process to setup an aircraft the sheer number of people that are using this autopilot IS THE MOST VALUABLE FEATURE! Not to mention its incredible flexibility and feature-set & camera Pan-Tilt support :)

I found a fellow Paparazzi member AntonK living close to me and he has helped me out GREATLY!

I will aggree their are some nice and Simple autopilots out their and they are cheaper and I fully support the KISS principal. My recomendation is if you go with one of the other autopilots find somone who has a successfull implementation of that autopilot and copy the setup identically, and even try and contact them to see if they will help you recreate their success.

2 months, I started one month ago and I can see my self finishing in another month actually would be sooner if I did not have to send my GPS/IMU into be updated, Im not using the IR board. I have spent $800 not including my IMU/GPS module. (I did not buy my RF modem from PPZ UAV as I wanted the new 50mw 2.4ghz zigbee pro)

I recommend you speak with David(BMW) from PPZ UAV. He produces the hardware and is a very helpful person!

Atto Pilot I must say is very promising and I would highly recommend you speak with the creator...

PPZ is the oldest most tested and used autopilot I would assume? (Maybe someone can correct me?) I even venture to say the PPZ is more tested and used than some of the commercial Autopilots. I have heard stories of people swapping out Micropilots/with a PPZ system at the last hour just to save a failing UAV demonstration and the PPZ worked right off the bat. (granted these people had experience with PPZ)

In the end the community support for PPZ is what sold me and a group of 10 engineering/science students.
So in my opinion the level of support is the most important element.

Good Luck!
Scott

[rich smith]

I must agree with Scotts assesment and plead guilty to a little trolling. It's hard to resist getting a rise out of the Linux dudes and Paparazzi too. Almost always results in an eductional experience though.

Paparazzi probably was the first DIY autopilot effort and I certainly blame them for dragging me into this. The Mega8 in my current autopilot was inspired by their use of this chip back then. At first I thought my difficultes with early Paparazzi were due to "bleeding edge" growing pains but browsing recent threads here on RCG leads me to believe things haven't changed much.

IMO you can't get one going without help. Many of these type projects are more of a "cult of personality" than technical effort. Usually one person, charismatic and hyper-motivated, drives the enthusiasm with a small number of socially bonded acolytes and large or small gagle of followers. Sometimes they accomplish more than governments or teams of scientists simply because of this semi-religious fervor.

Problem is lack of time and resources. Like Scott, my motto is KISS and I'm famous for taking shortcuts. Then there's the fact that you need to attach to one of the Gurus to succeed and I never did get high marks in "Works Well With Others". It also seems an individual can sometimes get 'er done with less effort and in shorter time than design by committee. :)






Paparazzi is NOT!!! strictly for DIY masochists, but Rich sure has a good point. Its a little daunting if you don't have help from the community.

If you get involved in the PPZ community you will see that while it is a complex autopilot and is not a 123 step process to setup an aircraft the sheer number of people that are using this autopilot IS THE MOST VALUABLE FEATURE! Not to mention its incredible flexibility and feature-set & camera Pan-Tilt support :)

I found a fellow Paparazzi member AntonK living close to me and he has helped me out GREATLY!

I will aggree their are some nice and Simple autopilots out their and they are cheaper and I fully support the KISS principal. My recomendation is if you go with one of the other autopilots find somone who has a successfull implementation of that autopilot and copy the setup identically, and even try and contact them to see if they will help you recreate their success.

2 months, I started one month ago and I can see my self finishing in another month actually would be sooner if I did not have to send my GPS/IMU into be updated, Im not using the IR board. I have spent $800 not including my IMU/GPS module. (I did not buy my RF modem from PPZ UAV as I wanted the new 50mw 2.4ghz zigbee pro)

I recommend you speak with David(BMW) from PPZ UAV. He produces the hardware and is a very helpful person!

Atto Pilot I must say is very promising and I would highly recommend you speak with the creator...

PPZ is the oldest most tested and used autopilot I would assume? (Maybe someone can correct me?) I even venture to say the PPZ is more tested and used than some of the commercial Autopilots. I have heard stories of people swapping out Micropilots/with a PPZ system at the last hour just to save a failing UAV demonstration and the PPZ worked right off the bat. (granted these people had experience with PPZ)

In the end the community support for PPZ is what sold me and a group of 10 engineering/science students.
So in my opinion the level of support is the most important element.

Good Luck!
Scott

[mecevans]

well ive seen the ppzuav site and i can get a complete http://ppzuav.com/osc/catalog/product_info.php?cPath=2&products_id=68 (TINY for $655 and a http://ppzuav.com/osc/catalog/product_info.php?cPath=2&products_id=88 (TWOG for $700. they are now even including pre-crimp servo wires, which i know where a pain in the past due to the $100 crimpers.

[airmcn_3]

well ive seen the ppzuav site and i can get a complete http://ppzuav.com/osc/catalog/product_info.php?cPath=2&products_id=68 (TINY for $655 and a http://ppzuav.com/osc/catalog/product_info.php?cPath=2&products_id=88 (TWOG for $700. they are now even including pre-crimp servo wires, which i know where a pain in the past due to the $100 crimpers.


$100 more and you have a very proved and well tested attopilot. I can tell you from experience as I have just over 200 flights with Attopilot. Once Dean comes back from his trip and gives me permission I will post accuracy pics.

Chris

[mecevans]

what will be included in the kit? i dont see where you can purchase them yet,are you testing out one of the production samples or something?

[airmcn_3]

what will be included in the kit? i dont see where you can purchase them yet,are you testing out one of the production samples or something?


The kit includes an attopilot unit, XY sensor, Z sensor, Micro SD socket, Voltage monitor, Telemetry interface, GPS head and the wires to connect it all. Yes I am testing as hardware and firmware updates are made.

I am not really trying to sell you on it just wanted you to know that you are close to the price range with the kits you mentioned above. It’s a very good autopilot, extremely accurate and has tons of user SET options.

I can chat with Dean and see what he says about contact info in regards to purchasing an attopilot if you like.

Chris

[mecevans]

seems like a good product, are there no modems included?

[airmcn_3]

seems like a good product, are there no modems included?


No you need to supply your modem. Most of us are using 9Xtend 900 MHz but you can use what ever you like.

Good luck with your project. It’s a very satisfying feeling the first time you get it to fly itself.

[mecevans]

hmm so thats going to bring it into the 1k plus range. i still need a little money left over for a scratch build/ arf and a motor and stuff.

[mhaun5]

Chris,

Dean is standing next to me, and says go ahead and post your flight results.

Mecevans,

Contact Dean for student pricing and availability.

Thanks,
Mike

[mecevans]

awesome!! thanks! :)

[airmcn_3]

Chris,

Dean is standing next to me, and says go ahead and post your flight results.

Mecevans,

Contact Dean for student pricing and availability.

Thanks,
Mike


Mike,

Groovy,

These pics are of a flight not too long ago. This was done with multiple spotters on the ground across the complete flight path we had communication between each individual.

The line hold was +or- 1m most of the flight, there were times through turns and the occasional wind gust that I saw up to 5m. Altitude was +or- 2m, again dependent on the wind gusts. The curvy part in the right corner is me in manual mode, the circle is in holding pattern.

Its late here in OKC so I will post charts tomorrow.

Chris

[airmcn_3]

hmm so thats going to bring it into the 1k plus range. i still need a little money left over for a scratch build/ arf and a motor and stuff.

You do not need the modem for auto flight, it is just a added bonus to have GE tracking and live data. Atto does not even require a Laptop at the field to fly auto flights.

[Buzz_Lightyear]

I would refute the claims that one needs a "guru's" help to fly paparazzi. The total for my parts were:

$400 for the autopilot part (autopilot, IR sensors, cables)
~$150 for the RC stuff (airframe, radio, receiver, though I did have some parts already)
$350 for two 9XTend modems

That puts my complete system flying system right around $900. I had this system flying in two months, almost entirely on my own. The only help I had was tracking down an issue with a non-functioning IR sensor. Once that was repaired, following the wiki got me flying in a couple of weeks.

It's true that Paparazzi can seem daunting at first because there are so many available options and it is so customizable. But once you take some time to read through everything and begin to comprehend it, you start to see just how amazing the project is. Not to mention the community. I would highly recommend the project.

* Disclaimer - I haven't used neither ardu nor atto. But I have used a few other commercial autopilots.

[ios]

Hi Chris :)

I can't believe you've had the chance to have 200 flight with Attopilot already, thats awasome.

Your GoogleEarth pictures of the flights are fantastic, and it really excites me to know whats just around the corner. I've attached here one of my own very successful flights I plotted on GoogleEarth late last year.

Take Care

[Gary Mortimer]

I am in no way at all jealous of the guys in the USA that have early Atto units in their sticky paws.

No me never no, honest really, no.

No not even slightly.

Seems like Airmcn has been a big part of straightening any wrinkles and along with the lucky few that are now receiving units, once export restrictions are lifted us overseas types can look forward to a fully wrung out system.

Can't you twist Deans arm Airmcn to let you post some more stories in a new thread of your experiences so far???

I'm sure there are lots of interested eyes here.

Cheers

G

[airmcn_3]

I am in no way at all jealous of the guys in the USA that have early Atto units in their sticky paws.

No me never no, honest really, no.

No not even slightly.

Seems like Airmcn has been a big part of straightening any wrinkles and along with the lucky few that are now receiving units, once export restrictions are lifted us overseas types can look forward to a fully wrung out system.

Can't you twist Deans arm Airmcn to let you post some more stories in a new thread of your experiences so far???

I'm sure there are lots of interested eyes here.

Cheers

G


Gary,

I am sure I can twist Dean’s arm a bit, I will be going to his place on Monday for a couple of days. It’s true, I have a lot of flights on Atto although the only thing I do is test and suggest.

Dean has a very good grasp on how this autopilot thing needs to work; he is one incredibly smart dude.

I can give you guys some tips on tuning once you have received your units. The tuning of an aircraft with atto is going to be rather simple with the new live PID tuning that has been introduced. One thing to be aware of is you will need a modem in order to do live tuning.

Chris

[airmcn_3]

Hi Chris :)

I can't believe you've had the chance to have 200 flight with Attopilot already, thats awasome.

Your GoogleEarth pictures of the flights are fantastic, and it really excites me to know whats just around the corner. I've attached here one of my own very successful flights I plotted on GoogleEarth late last year.

Take Care


Already is a bit understated, I have been working with Dean for over 5 months now so 200 flights is not that hard given I do this for a living. If the weather was par more often that number would be much higher ;)

Nice flights you have attached. What system were you using?

Chris

[Gary Mortimer]

So give us a clue Chris, do you hand fly the aircraft.

Find some base data using the PID charts.

Then season to taste??

With the modem, I'm guessing you can change that on the fly, watch what happens and then accept the best settings??

Can the data then be crunched over many flights to generate settings that you were never smart enough to work out??

Sorry going off topic for this thread.

Hurry up and open an atto operational thread you atto chaps

G

[airmcn_3]

So give us a clue Chris, do you hand fly the aircraft.

Find some base data using the PID charts.

Then season to taste??

With the modem, I'm guessing you can change that on the fly, watch what happens and then accept the best settings??

Can the data then be crunched over many flights to generate settings that you were never smart enough to work out??

Sorry going off topic for this thread.

Hurry up and open an atto operational thread you atto chaps

G

Gary,

Yes the first thing we do is fly in manual mode, we collect all the data and study it. The real work is in ARC mode, this is when the autopilot is working with all the sensors, once you have ARC flying nice and smooth you can begin auto flights. The one thing you will not have rung out will be the navigation tuning; we have a very good idea where they need to be given the amount of different airframes we have flown atto on.

It’s really a quick flight, turn to auto, watch the habits and bring her home, you then study the data and adjust accordingly.

The live PID tuning will be just as you described, you can move the sliders around until the aircraft is tamed to you liking.

If you are a Excel guru you can take the multiple files and mesh them together for a good amount of data to study. I have found nothing gets you closer then real world testing as the data can make your eyes go cross eyed after a bit.

I will try and contact Dean this afternoon and see what he says about an Atto user thread.


Thanks,

Chris

[mecevans]

Wow! thanks for all the information! it looks like it holds the waypoints perfectly! how many flights did it take you to tune it to your liking? what is the ground station like? can you adjust the flight plan in flight? does it have the loiter and survey functions like paparazzi?

[workshop]

I'm looking forward to that!

Without a way to visualize the huge amounts of data that Atto collects, the "bring her home, analyze the data, tweak, and yer done" method isn't working for me.

I'm using the Catalina 1 test plane that has been home to three autopilots and I'm having a hard time visualizing all the variables needed to fly right. Without those 200 flights, it is hard to tell what variable cures a symptom. Serves me right for watching the deveopment process from afar but that makes me a typical "first time" user.

It would take some pressure off Dean if we had a thread or even a RCGroups vendor forum...

Jeff

[airmcn_3]

Wow! thanks for all the information! it looks like it holds the waypoints perfectly! how many flights did it take you to tune it to your liking? what is the ground station like? can you adjust the flight plan in flight? does it have the loiter and survey functions like paparazzi?

It took 8 flights to get this particular aircraft where it is now. The GS is nice, you can get a good idea of it a few pages back in the Atto thread. Yes you can adjust flight plan in flight. Yes it does have the loiter function as well.

[airmcn_3]

I'm looking forward to that!

Without a way to visualize the huge amounts of data that Atto collects, the "bring her home, analyze the data, tweak, and yer done" method isn't working for me.

I'm using the Catalina 1 test plane that has been home to three autopilots and I'm having a hard time visualizing all the variables needed to fly right. Without those 200 flights, it is hard to tell what variable cures a symptom. Serves me right for watching the deveopment process from afar but that makes me a typical "first time" user.

It would take some pressure off Dean if we had a thread or even a RCGroups vendor forum...

Jeff


Defiantly not as easy as bring her home analyze the date and done, but you can get a lot of data out of short flights. The short flights are just to get your rough parameters down, after that there is a considerable amount of data that needs to be studied to get it perfect. Each airplane is different. The symptoms are rather easy to recognize, Dean does a very good job of explaining each SET parameter in the manual if you do not have a UAV background. I assure you it is not too hard.

Chris

[workshop]

I've been flying RC for about 12 years. I have about 4-5 years flying UAVs 1 to 12 miles beyond visual. I have logged 40Hrs. flying FPV.

4 flights on Atto and I'm still all over the sky. Looking back, PicoPilot only had two pots to tweak. I'm hoping the added complexity and time to tune is rewarded by a very robust AP.

Jeff

p.s. The "first time" use comment refered to AttoPilot

[mecevans]

NICE! seems like everything i would need! i hope i can get one.

[airmcn_3]

I've been flying RC for about 12 years. I have about 4-5 years flying UAVs 1 to 12 miles beyond visual. I have logged 40Hrs. flying FPV.

4 flights on Atto and I'm still all over the sky. Looking back, PicoPilot only had two pots to tweak. I'm hoping the added complexity and time to tune is rewarded by a very robust AP.

Jeff

p.s. The "first time" use comment refered to AttoPilot


Got ya! Not a bad background!

The added complexity only makes it better :D If you have any questions fell free to PM me I will do the best I can to help you out if you like.

Chris

[mecevans]

whats the best was to contact dean? i need to start getting a list of all the parts i need together, and come up with an idea on an airframe. what are the dimensions of the attopilot?

[mhaun5]

whats the best was to contact dean? i need to start getting a list of all the parts i need together, and come up with an idea on an airframe. what are the dimensions of the attopilot?

The best way to contact Dean would be to visit his website:

www.attopilot.com

Mike

[mecevans]

i have not heard from Dean so i went ahead and ordered the complete tiny 2.11 system. it should be here pretty quick since i live close David (thanks for answering all my questions :D ). i have a optic 6 radio and it has a 3 position switch so i should be able to use that. my next step is coming up with ideas on airframes with about a $300 budget, funding will be increased further if i can sell all my heli stuff. thanks for all the help you guys are giving me so far, i greatly appreciate it! :)

[mhaun5]

i have not heard from Dean so i went ahead and ordered the complete tiny 2.11 system. it should be here pretty quick since i live close David (thanks for answering all my questions :D ). i have a optic 6 radio and it has a 3 position switch so i should be able to use that. my next step is coming up with ideas on airframes with about a $300 budget, funding will be increased further if i can sell all my heli stuff. thanks for all the help you guys are giving me so far, i greatly appreciate it! :)

Good luck with your project. Make sure you create a thread so people can track your progress.

Dean has been out of town (with me actually) working on installing the product with my airplane; hence the reason he hasn't been able to respond to you yet. We had great success in getting the plane going very quickly with the AttoPilot unit (once the plane was ready, the AttoPilot was ready to go within an hour up in the air!).

Hopefully you can have similar success, sooner than BMW330i has - for his building thread has been going on for 300 days now!!! :D (Fun innocent poke @ David... Its encouragement to get his butt outside and start flying!)

[mhaun5]

I've been flying RC for about 12 years. I have about 4-5 years flying UAVs 1 to 12 miles beyond visual. I have logged 40Hrs. flying FPV.

4 flights on Atto and I'm still all over the sky. Looking back, PicoPilot only had two pots to tweak. I'm hoping the added complexity and time to tune is rewarded by a very robust AP.

Jeff

p.s. The "first time" use comment refered to AttoPilot

Jeff, I am surprised you are having issues; as mine flew out of the box near perfect. I have made a few adjustments to tighten a few things, but I couldn't be pleased more.

If you like, shoot me a PM and I can try to help you out. I have had 6 flights on mine now, and it works great!

Mike

[mecevans]

I am considering a design like IAD's current paparazzi airframe. he claims to get an hour of flight with 2100-3000 mah batteries!

http://www.rcgroups.com/forums/showpost.php?p=10769608&postcount=714

http://www.rcgroups.com/forums/showpost.php?p=10276474&postcount=362

[rich smith]

That is entirely possible. I've gotten over 100 minutes on 2200mah. You cannot use construction techniques with total disregard for weight as is common around here though. And "SUV" power systems are out of the question too.

I am considering a design like IAD's current paparazzi airframe. he claims to get an hour of flight with 2100-3000 mah batteries!

http://www.rcgroups.com/forums/showpost.php?p=10769608&postcount=714

http://www.rcgroups.com/forums/showpost.php?p=10276474&postcount=362

[mecevans]

i have started my own paparazzi how-to / build thread, check it out.

http://www.rcgroups.com/forums/showthread.php?t=1000937