If you're in the San Francisco Bay Area, you can meet the ArduCopter team and see demos tomorrow at the East Bay Maker Faire in Oakland. Jason Short and I will be giving a presentation at 12:00 on the main stage. Quads will fly! Come by and say hi ;-)

A few days ago I posted some videos of my object tracking algorithm, and some folks were curious how well it would perform if the object left the frame. Since posting I have hit kind of a breakthrough in the algorithm and it is working much better than I had ever expected. I am still fine tuning some of the sensitivities, but I added some neat features.

1. Now you get a white box when you click and drag to define the target.
2. There is an adjustable accuracy threshold. If one finds that the target will not re-aquire, one can reduce the accuracy (demonstrated in the video above)
3. New improved multi-threading (under the covers)

I am working on both Mission Planner integration and, for those who want to use this for auto antenna tracking or camera tracking from the ground, Pololu Micro Maestro integration. As you can see in the video above, the software is able to re-aquire the target when it leaves the frame or gets confused. This is thanks to the constant learning of that objects features as it changes shape or orientation over time.

Finally, someone has created a PID tunning howto video, from scratch to perfect stabilize flight, that I think will be very useful for the arducopter community, specially for those flying non stock frames.

The concepts are already in the AC2 PID tweaking guide developed by Jason, but a visual howto helps a lot!

The guide has been created by Menno (joebarteam), an active user of the megapirates community in rcgroups. Full credit to him, I'm just spreading his post.

First, acro mode tunning (which I always discard, but I'm realizing it is pretty relevant):

Second, stable mode tunning:

The full post with Menno comments is here.

Note:

If the arducopter developers think it is useful maybe it could be added to the wiki (with Menno permission obviously).

CNN did an episode on DIY Drones today. Good shots from the back of 3D Robotics, including clips of Jeff Taylor and Gigio Romero (Udrones) flying ArduPlane and ArduCopter.

The above is a plot of the 3 components of estimated magnetometer offsets computed during spin tests at 78 RPM of an improved method for estimating and removing magnetometer offsets. There is a report available with theory and implementation. The method will work equally well for fixed wing aircraft, multicopters, and helicopters. It makes no assumptions about the dynamics of the airframe.

The idea for the new method came to me while I was recently working on a method to detect and compensate for magnetometer misalignment errors. I realized the method that I had been using for offset compensation was sensitive to misalignment between the magnetometer axes and the gyro axes, so I thought about ways to compute offsets that would ignore misalignment. I found a good way to do it that turned out to be more accurate and easier to implement than the method that I was using.

I also figured out a slick way to detect and adjust for magnetometer misalignment, I will be publishing a report in a few days on that subject. The method will compensate in flight for any amount of magnetometer misalignment, including a 180 degree yaw mounting error. In other words, the algorithm can figure out that you mounted your magnetometer backwards and apply a rotation matrix to the magnetometer vectors to put them into the correct reference frame. Stay tuned....

[Here is the technique for doing inflight magnetometer alignment. - WJP]

Best regards,

Bill Premerlani

Thanks to Chris, according to many who had been at Number Twelve yesteraday was a very interesting meetup. One of the rare one's where you talk about everything from UAV's to how UAV's could be used to monitor eggs in bird's nest to export regulation of UAV's.

My daughters and I have built several quadrotors out of aluminum and also a mini quad out of kite parts (see beatty-robotics.com for these). But for this next project, we decided to make a quadrotor out of carbon fiber. We applied everything we learned on our previous builds and also tried a few new ideas that we hadn't tried before. 

The arms of this new quad are made out of (4) 8" lengths of 3/4" square carbon fiber tube. On our past builds, the motor mounts often bent (and had to be replaced) on "hard landings," so this time we dispensed with the motor mounts altogether. We drilled the mounting holes directly into the arms (as shown in the picture below). That may be worse. It might just break the arm instead of bend the motor mount (which would be a worse failure mode obviously). We'll see how it goes. I noticed recently that the 3DR kit also appears to use the same technique on their aluminum arms.

The center plates are made out of two 6" x 6" squares of carbon fiber. In previous builds, we always stacked multiple plates vertically to mount all the electronics and store all the gear. In this case we decided to "go wide" instead of "go tall."  The idea was to flatten it out and make it more crash resistant, and to reduce overall weight and parts by having fewer plates, fewer fasteners, and no standoffs at all (the arms separate the plates). It also provides more of a fixture-surface for the arms and it gives us a place to store the ESCs and all the wires (We solder our own Power Distribution Unit out of Deans connectors). We'll see how that works.

Next, we wanted to try the switch idea that I had discussed and asked about in the forum. We have installed a decent-sized (but relatively light) illuminated automotive toggle switch. When the quad is stored in the house, the battery will be disconnected, but out in the field, we'll use the switch to turn the quad on and off. It's very quick, easy, obvious (red), and it can be reached without putting your hands in the path of the props. We're hoping this will be convenient and safe. In retrospect, it might have been even better (safety wise) to put the switch on the outside edge of the underside. (Note: I'm aware that 20 amp automotive switch may or may not be suitable for the ultra-high amps of a quad rotor, but refer this Forum post for our extensive discussion on that subject). We'll see how it goes.

We've had a lot of trouble with landing gear in the past. We've tried a number of approaches and none of them have been satisfactory for us. On this build, at least initially, we decided to test out the 3DR landing gear pieces available in the DIY-Drones store. They seem like they are going to work well. 

I have uploaded pictures of the completed quad as well as an image of our CAD drawing and a few in-process pics. It's too wet to fly today, but hopefully we will be able to post a video soon. 

I thought i would share my hexacopter build with the world, it's not finished as of yet but it is getting there.

This is my second multicopter build; the first being a quadcopter which i may make a build log about as well but basically it was to see if my choice of building materials etc would hold up with the hexacopter project.

Components I am using:

- NTM Prop Drive 35-36A series (1800k/v @ 875W)

- HobbyKing Redbrick 80amp esc

- Ardupilot Mega board

          - Megnetometer

          - Xbee Pro Telemetry

          - GPS Module

- Spektrum DX8 tx/rx

- 10x4.5 props

- Carbon Fibre square tube 6mm

- Carbon Fibre sheets 300x300mm

- M3 nuts,bolts etc

- 3Amps @ 5v Micro UBEC

- Turnigy 5Ah 3cell (11.1V) LiPo (x2) (havent purchased yet)

- Assortment of silicon wire (24awg, 14awg, 10awg, 4awg)

- Assortment of plugs etc

Thought i'd better edit the post and add some of the first pictures i had taken...

Now only if i had access to a laser cutting or CNC machine... one can dream i guess. Main chassis plates:

Motor "mounts". Yes you may say i did a poor job of the mounts (not very well designed) but i wanted a light mount (so i could keep frame weight to a minimum) and something nice looking. I didn't go the modular approach with any of the frame design as i didnt see the point in my circumstance. to change the motor incase one breaks there are 4 phillips head screws i can undo to get it off. The roughly cut x-brace under the motor was to see how well it would work with a brace there. The gap between the 'shafts' on the arms is taller now and the x-brace is nicer to look at ;)

Getting ready for the assembly of the frame :)

Bog roll there was to wipe off excess locktight and i was sick at that stage and we had run out of tissues...

A third of the way there... thinking this thing is going to be huge. Always looks bigger off paper

Mainframe assembly complete. motors attached and escs placed where i think they should go.

Comparison next to my quadcopter:

Power distribution looms:

I will upload more pictures as i make progress. i will receive a new esc next week as well as the rest of the components i'm waiting on.

To do:

- Battery tray (another sheet of carbon fibre (hexagonal in shape) that will be suspended underneath the mainframe centre)

- Landing gear (not sure on how to tackle this yet)

- Connect electronics

- Find a method to mount the Arducopter Mega in between main frame CF sheets (throught the hole in the top sheet)

- Make up some brackets to mount GPS, Xbee Pro, Rx onto

- Probably numerous other small things

I would put down weights and stuff but as i havent finished the build it is a bit hard to say, but weights so far and projected weights are:

- Mainframe with escs, motors and all heavy wiring = 2100 grams

- Mainframe by itself = 600 grams

- Battery weight = 860 grams (10Ah)

So total flying weight im hopeing for is 3000 grams but i've given myself a worst case senario of 3500 grams.

Total lift with 10x4.5 props = 17.4kg (i was hopeing for more but because they are such high rpm motors the best props i wasn't able to find in both puller and pusher)

So worst case senario power:weight is around 5:1 so i could get around a 10kg maximum payload i guess (still gives me a 2:1 power to weight); I would have to wait and see about that though.

I tested power consumption of the motors with around 3.5kg of weight and i calculated i will have around 15 to 20 minutes flight time without a payload and i have made enough room for 20Ah of batteries if i want that bit extra flight time.

So far im happy with the frame and its strength. it takes considerable force to even move the arms and bend it.

If you want any pictures taken of it and uploaded i'm happy to do so and any feedback would be greatly appreciated and if you have any questions i will do my best to answer them as soon as i can.

Regards,

Jamie Good

-------------------------------------------------------------------------------------------------------------------

UPDATE:

So after a couple days of work and study i have had a part day to work on the hexacopter. Today i had my first flight (didn't go above 1 metre from the ground). There will be a couple hours of PID 'tuning' to do to get it to fly nicely. At the moment hover is at about 35% throttle (thats with a linear throttle curve). My hover power consumption estimations were a bit out as in hover the maximum i recorded was 54 amps (instead of my initial prediction of 40 or so amps) so my flight time with 10Ah of batteries wont be what i predicted.

The main components adding greatly to the weight I still have to make/install are landing gear which i hope to do as lightly as possible and i need to add another CF 'layer' to hold the batteries (unless i figure out another nice,clean way to hold batteries). so im hoping i will only have to add another 300 grams or so... fingers crossed.

I was wanting a video to post now but as my partner wasn't home and i dont have a proper camcoder etc (have to use our phones) i will need to wait for her to get home. Hopefully my batteries will be charged and it will still be light by that time so im hoping i can upload one tonight.

If anyone knows of an easy way to tune the PID values could you please let me know? as of now i am flying, landing, adjusting (up or down as needed) which is very time consuming and annoying.

Thankyou,

Jamie

-------------------------------------------------------------------------------------------------------------------

UPDATE:

Second ever flight. I changed the PID values a little to make for a 'smoother' flight... but they are a long way off from where they should be. Again if anyone knows an easy way to tune them it would be awesome if you share them.

Hopefully i can get these settings calibrated and frame finished so i can upload a proper flight video for everyone. Can;t wait to get a camera on it!

-------------------------------------------------------------------------------------------------------------------

UPDATE:

Took the copter out again today to show a couple mates. Still haven't touched the PID values but i plan on doing some tonight after learning that i can adjust each value via channel 6 (the pot) on my Tx, which should make things easier. All components are still sitting ontop of the chassis which isn't ideal as the IMU isn't level with the copter but i can just compensate with the controls :P.

Nice windy day at the beach :).

"Can it land on the table?"

"I Guess..."

Flying in a Garage... dont you just love Prop wash?

Also here are some pictures taken from the day. Please excuse the mess sitting ontop of the chassis :)

After a couple minutes of tidying the electronics i have a couple things where they should be...

Also while i'm thinking about it, i need a way to tell the orientation of the thing while its up in the air... i was thinking maybe some fluro orange tape or paint and make an arrow or line underneath a couple of the arms... any ideas?

Here is a short clip where you can see HD video that was captured from a flight of my Creep 1.0 Tricopter.

Mad Scientists

I know it was already touched on earlier but I saw this on yahoo news and wanted to share...

A computer virus that hit the US drone fleet last month created a "nuisance" but no serious threat to flight operations for theunmanned aircraft, the US Air Force said.

Confirming the virus for the first time days after it was first reported, Air Force Space Command said computers at a Nevada air base were infected with malware but played down the problem as a minor headache.

Malware on the computers at Creech Air Force base, where drone planes over Afghanistan and elsewhere are piloted remotely, was first spotted on September 15, Space Command said in a statement.

"The infected computers were part of the ground control system that supports RPA (remotely piloted aircraft) operations," it said.

"The ground system is separate from the flight control system Air Force pilots use to fly the aircraft remotely; the ability of the RPA pilots to safely fly these aircraft remained secure throughout the incident," it said.

The malware, of a type that is "found routinely on computer networks and is considered more of a nuisance than an operational threat," had been isolated, it said.

Wired magazine first reported the virus last week, saying it had spread at Creech through removable hard drives used to load map updates and transfer mission videos from one computer to another.

Drone units at other US bases around the world have now been told to stop using the removable hard drives, the magazine reported.

A spokeswoman for Space Command said that rules prohibiting openly discussing the operational status of aircraft were waived to assure the public about the state of the drone fleet.

The Air Force "felt it important to declassify portions of the information associated with this event to ensure the public understands that the detected and quarantined virus posed no threat to our operational mission and that control of our remotely piloted aircraft was never in question," Colonel Kathleen Cook said in the statement.

The American military and intelligence agencies have come to heavily rely on the robotic aircraft to track enemy forces, with the planes transforming battlefield tactics and strategy in recent years.

The US military openly acknowledges drone flights in wars in Afghanistan, Iraq and Libya but the CIA declines to publicly discuss its covert missions using drones to take out Al-Qaeda extremists in Pakistan, Yemen and elsewhere.

http://news.yahoo.com/us-air-force-calls-drone-fleet-virus-nuisance-053626091.html

You're probably familiar with Al Shan Mao who vblogs on RC topics from China.  He's got a new video up previewing a UAV he will be receiving and reviewing.  Does anybody know anything about this unit?

Highlights:

• full autopilot and GCS
• 4 years of development
• from "China's leading UAV makers"
• interesting elastic-propelled takeoff and parachute landing systems
• video feed with target focusing

APM_configs.xls

Hi all pls find the actual version of the recommended configurations from here

http://code.google.com/p/ardupilot-mega/downloads/list

What is intriguing is that

XTRK_GAIN_SC is varying widely, very high PTCH2SRV_P for skywalker,

and unexpected differences in rudd tuning for bixler and easystar.

What is your favorite 'best setup'? Please share.

Who am I kidding right? its almost impossible to design an indestructible quad copter frame! That's why this frame is actually the complete opposite. Its designed to fail! 

This is nothing new. Most designs nowadays are designed to fail, but they are designed to fail correctly and safely. There is always something that needs to be protected and for that to happen something needs to fail. Like crumple zones in cars or pressure relief valves on pressure vessels.

In the case of a quadcopter, I'v got not passengers to look after but I would like certain parts (the expensive parts) to survive crashes. Also after crashes it should be easy to fix.

Iv taken what Iv learnt from my arducopter v1 frame (which is now very tired) and my engineering diploma and put together a frame that will protect all the valuable items and break the cheap readily available parts.

So, every nut and bolt is nylon except the motor mounts. Any hard impacts will shear the bolts and save the frame. This happened alot with my arducopter frame. its a good thing =) rather break a bolt or 2 than a landing gear. I will be placing some high density foam right in the center of the frame for when an arm gets pushed into the frame. This should absorb some energy and stop further damage as those carbon booms dont break easily. I have placed the motor mounts slightly inside the boom as I think they are vulnerable at the ends. Yes, this now makes booms vulnerable but nothing a rubber stopper cant fix =) Where the arms meet the frame and motor mount I want to put thin strips of high density foam to absorb some vibration from the motors and also absorb some energy in crashes.  

So hopefully all I will be replacing after crashes is nylon bolts. Even if not, I still like the frame =)

More about the frame itself:

The arms will be shortened 450 heli carbon tail booms from HK, $2.40 link

The carbon sheet is 1mm thick also from HK which I will cut with a dremel, $10.79 link

landing gear will be 3mm clear Perspex strips bent will mini blow torch

Motor mounts will be 4mm perspex discs made with a hole saw and file =)

Electronics, motors, etc will be from my arducopter frame.

Will post a build blog when its done.

Please criticize =)

Hi everybody,

I am starting with new hexacopter Project. It is my own project.. I am totaly beginer

an I hope, that project will be succesfull.

The frame is made from plywood 3mm, cuted with laser.

Last week, I ordered APM from jDrones.. I am still waiting postman.. :-) APM is on the way (on the air I hope) be cause I am from middle of Europe.

Motors will be AC2836-358 with 12X45 propeller set.

Plywood will be finished with polyester resin and waterresist polish.

My progress you can watch on this blog :-)  .. so I will be pleased for every advice and comment...

-jan-

The German effects studio OMStudios must really trust their custom RC Octocopter--the super high-def RED camera they're lifting costs $40,0000.

Via Engagdet

I'm just getting all my equipment sorted out for FPV on the quad, but made some good progress today that I thought I'd share.

I got the idea for a standalone FPV pod from flitetest.  Just wanted to get things working quick and dirty for both the quad and the EZ*

Wrapped up in there I have the following:

1)  Camera is 420TVL from here for $32 (free shipping when you order 2!)  The cameras are easily removed, and seem to be good enough to start with.

2)  Tx/Rx is 5.8Ghz from HK for $63

3)  Bluebeam antennas from here $65 for the pair unless you are capable of making them yourself

4)  Don't forget the adapters for this antenna VTx/Rx combination from here $10 for two

5)  It also has a 3S 500mAh battery like this $12 - I plan on dropping this and creating a single battery setup

Strange, I just hit "save as draft" didn't mean to post this yet, but here's the video anyway!

I am currently working on integrating an object tracking algorithm of my own making into the Mission Planner. The algorithm combines Lucas-Kanade and SURF and attempts to learn negative and positive reinforcement values as it runs. Right now it is in the early stages of development, but I hope to have something fully functional within a few months.

My hope is to use this to aid in automatic centering of the tracked object by sending commands to the APM to move the camera gimbal based on the distance of the tracked object from the center of the frame. In the videos below you will see two demonstrations of this tracking algorithm. In both cases it seems to perform well with background noise.

NOTE: I apologize for the terrible frame rate of the videos. I don't have very good screen capture software. The second video is a little higher quality.

After a few rough landings (cough crashes) and subsequent repairs, I got the feel for flying FPV. Canadian thanksgiving was this weekend, and there were still some nice colours in the trees. Just thought i'd share.

• I'm flying a Quadcopter running ArduCopter 2.0.45 with a Nikon Coolpix onboard for the video (want to get a goPro, but I had this lying around).
• 2.4 Ghz Radio
• 1280mhz video link
• 900mhz telemetry

Cheers,

Bart.

This is a port of Flytrons USB2PPM Project Compufly with enhancements 

You can use multiple usb sources and map any axis and button to any channel of your choice. 

You can also map multiple buttons to the same channel and adjust the ppm value to them for a 3 way switch with 3 buttons type set up

Current feature list

Servo/PPM output

1. Ability to output no less then 10 ppm Channels
2. Channels need to be able to be inverted
3. Channels need to be able to have Trims rate settings
4. Channels need to be able to have Exponential rate settings
5. Channels need to have multiple input to output mappings
6. Channel Resolution 2048 (11Bit)

Joystick/Gamepad Control

1. Accept 3 or more USB device inputs
2. USB button detection on joystick interface
3. Support for POV HAT Buttons
4. Support for HAT Switches
5. Tabs or Pages for all detected USB interfaces
6. Mapped Values for all axis for specific device per Tab/Page

Output Device Configuration

1. Selection of number of channels
2. Mode selection Positive/Negative PPM

GUI

1. Resizable for use on note/netbooks
2. Ability to save/load settings
3. Show values for ppm output and joystick/mini joy axis with Progress Bar type scroll

Everything is kind of setup for my CH Products set up with joystick throttle and rudder at the moment.

Check out the google code page and try it out for yourself. Make sure to read the wiki pages fore more info. 

Google Code: Joystick to PPM
I am looking for Java programmers and testers to help work out bugs. If you think you fit the build email me and I'll add you to the list. 

Screen Shots

aircool00 at g mail dot com

Cheers,

Ryan