Posted by Andrew Fernie on January 4, 2012 at 2:06pm
I would like to fly with live video and goggles, but this makes it tough to make any changes on the mission planner. To avoid any need to take off the goggles I integrated a series of menus in the OSD.
My setup, as reported here, uses a second APM on the ground.It monitors the MAVLink data stream with an XBee, drives a Remzibi OSD on a second serial port, and supports an Ardustation display and keyboard as a terminal on a third serial port. To make the menu selection as simple as possible I added a quadrature encoder using three of the unused inputs on the Ardustation (nominally the analog inputs, but setup as digital). The A phase of the encoder drives an interrupt, and the resulting ISR examines the A and B phase states to determine which direction the encoder know is being turned. The encoder includes a switch (push the knob in) that is connected to the third input.
The first time the knob is pushed the top level menu is displayed on the top line of the OSD with four selections plus an "exit". Turn the knob to choose the menu item, then push to select. Each selection, in turn, has up to four available menu options. Choices are: 1. Symbology - select between four symbology formats 2. WP - read waypoints from the Ardupilot, set to first WP to restart the mission, next to skip a WP, or previous to move back one. 3. NAV - set Auto, RTL, and loiter modes (yes, easier to use the transmitter control, but I wanted to try out more of the MAVLink commands) 4. Timer - start, stop, and reset a stopwatch timer on the OSD.
The menus are turned off automatically after five seconds of no input.
I find that that this is a fairly easy way to control the system, and it can easily be expanded to include other commands.
All of the symbology is driven with the Remzibi "$M" command strings that are a bit limited in available characters - seems just to be letters, digits, and a couple of others. I can't use the leading and trailing character options for the strings as they take up more space. I would have preferred using something other than "X" to indicate which menu item is being selected, but that was all I could come up with. Hopefully the minimosd will have a more complete set of command strings, or be open source to allow me to improve the presentation.
You can find a video showing the system in operation here. I was in HIL mode so no satellite count, and the GPS time clock has random data (relies on a mod to the GPS library).
Latest code can be found here. It links with the Arduplane 2.27 libraries from the repository.
Here's a cool hack of the Parrot AR.Drone using the EZ Robot controller board (a PIC32 board with bluetooth and some very easy to use desktop software). It tracks the color red and chases him around the room!
Posted by s.v.itterzon on January 4, 2012 at 7:00am
Last year me and a friend of mine started building a quadcopter from scratch. We wanted to build the machine and the electronics ourselves. We experimentad a lot, but up until today we did not yet succeed in getting the bird in the air with the home build electronics. We keep having some trouble reading the data of one of the sensors, thus the PID loop doesn’t work properly. More information about out QR1 quadcopter can be found on QR1 quadcopter
We recently bought the Hobbyking control board for quadcopters, and mounted this on the QR1 frame. It flew good, and we both wanted a proper quadcopter for flying purposes only. We recon the Hobbyking control board with other KK microcopter firmware will do the job perfect. We wanted the quadcopter to be small, and started designing a mini quadcopter.
We used the same parts lots of other folks use in there quad, listed below: Motor: 18-11 2000kv Micro Brushless Outrunner (10g) ESC: Turnigy 6a Brushless Speed Controller Props:5x3 Propellers (Standard and Counter Rotating) (6pc) Prop adaptors: Prop adapter w/ Alu Cone M5x2mm shaft (Grub Screw Type) Turnigy nano-tech 850mah 2S 25~40C Lipo Pack Control board: ]HobbyKing Quadcopter Control Board V2.1 (Atmega168PA)
We ordered all the parts for an amazing 110 euro’s, you almost get it for free.
After deciding which parts to use, we started designing the little monster. We came up with a design that we could machine out of 1,5mm aluminium, for prototyping purposes. We plan on making the final design out of a carbon fiber sheets. The quadcopter is only 176x176mm and will have a RTF flying weight of about 325grams. We gave it the name QRM, meaning Quad Rotor Mini.
Below are some images of the prototype machined out of 1,5mm aluminium. The LiPo battery in the pictures is a 3S 800mAh battery, and will be replaced for a 2S 850mAh equivalent.
With the new fast prototyping tecnology is possible to develop new technology from the concept to reality . With high performance DIY electronics and 3D Design tools today is possible to obtain this kind of result.
This is our experience :
We use some 3D Pro as EdgeWorks or Solidworks and Opensource Tools as Bleender for design the frame.
After design we send our advanced project to a fast prototyping company as Shapeways
Then After some weeks is possible to recive the Frame .
After I put all the electronic inside the new frame I'm ready to doing the first flight the result of first experience was great , in only two hours we put all electronic inside the frame in the next days i would doing also some outdoor flight and put in front of quad a Gopro HD 2 cam for FPV :
After this experience I can say that we can fly with a frame made with 3D fast prototyping techniques.
The next step will be review of the original project with some upgrade as :
less material for have less price and less weight.
At the moment the weight is around 360 gr is yet good.
In order to fly with:
Multipilot32+NAVI FULL the frame is compatibile also with APM+OILPAN
4 Keda 20-22 Motors.
4 Hobbywing 30 A
4 10 '' propeller.
a 3S 4500 ma lipo
around 1250 gr
The main upgrade in our mind is more space for 10 '' propeller originally was developed only for 8 '' prop
possibility to change the tail if one broke. Better design for open and close cover on top of quad.
Following up on the success of my relay low battery warning leds I have expanded that code to include a complete lightshow and to be compatible with the newest ACM codes (2.1.1) and the APM1 hardware. APM2 hardware MAY work, but I don't have APM2 at this time, so no guarantees.
Credits to Bill Sanford and Max Levine whose work has been an inspiration and a basis for these leds.
1)Software side :
By using the agmatthews userhooks it is now much simpler to upgrade to a new firmware without having to change the code each time. You just have to copy 2 files and change setting in the APM_config.h. The needed files AND my config.h file (for version 2.1.1r5) are in the following file : U4eake_showleds.zip
Updated files for ArduCopter 2.5 (compile with arduino1.0 relax patch) : u4-showleds.zip These still work for ArduCopter 2.5.3.
Update files for ArduCopter 2.6 : Showleds_ACM2.6.zip Because motors were put in a library, every motor_armed in the showleds code had to be modified to motors.armed().Also you have to put
#define COPTER_LEDS 0
into your APM_Config.h file to disable the new copterleds, which are enabled by default.
You need to copy the usercode.pde and the usercode.h files into your arducopter directory (overwriting the files already there) and add the settings below to your APM_config.h. DO NOT just copy my APM_config.h, it is only included as an example to clarify the settings.
#define MOTOR_LEDS 0 // 0 = off, 1 = on #define SHOW_LEDS 1 // set to 1 to enable AN7-AN15 controlled led lightshow #define RELAY_LEDS 0 // set to 1 to use the old relay led low battery warning
The voltages can be adjusted to match your desires. If you have a 4S battery for example, you should put higher values, like 13.2V for the LOW_VOLTAGE. INPUT_VOLTAGE is the voltage that goes into the APM (usually the voltage from you ubec). VOLT_DIV_RATIO is an calibration value that you can adjust so that your reported battery voltage corresponds to the real battery voltage.
After compiling the code with arduino and uploading it to the APM board, you also need to set battery monitoring to option 3 : battery volts
That's it for the software modifications if you have a hexa or are happy with the patterns as they are ! On a quad or octo you may want to adjust the patterns a bit, but I'll explain that further down this blog (see Adjusting Patterns).
The code and blink patterns are optimised for use on my hexa (6 leds strips on the arms + 2 spare ones).
The leds will go on when motors are armed and out when disarmed. During flight, different flashing patterns can be selected with Ch7. The other use of Ch7 is not affected, so you can still use it for simple or recording waypoints. The leds will flash with the last pattern in this file when ch7 is high (recording waypoint or simple) When battery gets below the first threshold (MID_VOLTAGE = 10.6V), the ledshow is overridden and the leds will flash with following pattern : led 1,3,5 on/off and led 2,4,6 off/on. So if the first ones are off, the others are on and vice versa. When second threshold (LOW_VOLTAGE)is reached (9.9V) the leds will flash quicker.
2) Hardware :
The code uses the AN8 tot AN15 analog ports (portK) on the oilpan of APM1. In the picture you can see them labelled AN0-AN15. AN8 to AN15 are the ones on the right (4x2 pins)
You can either connect 1 led directly to each port (the ports can supply 40mA current) or you can use a ULN2803 darlington transistor chip to drive the leds according to the following shedule (thx to Max Levine for this schematic) :
This shedule is for a quad and uses a ULN2003. The ULN2803 has 8 ports instead of 7 and if I remember correctly it can supply a little more currrent (500mA per port). A typical hobbyking 12V ledstrip of 1m (60 leds) consumes about 300mA, so normally you should be able to hang 1m of ledstrip on each of the 8 ports. That should be plenty !!
You can get the ULN2803 in your local electronics shop or online here : Canadadrones - ULN2803
The darlington chip lets each input control the output directly across on the other side of the chip. So nr1 in the picture above, controls output 16, Nr 2 controls output 15 and so on. If you look at the pinout picture of the oilpan, you'll see that nr 1 is connected to AN8 and nr2 is connected to AN10, so AN8 in the pictures above controls the red leds, and AN10 controls the green leds. Get it ? ;-)
After the brainsurgery it will look something like this :
Now you just have to put a heatshrink over the ULN2803 chip and tuck all the wires away nicely.
CAREFULL : in the picture above I had soldered the wires to the wrong ports (side of the compass). That's what you get when working on your copter till 4 in the morning... It took me a day to figure out why my leds didn't work as expected. So you have to solder your wires on the OTHER SIDE (side of the usb connector). I didn't have a picture of when I soldered it right.
Finally you'll also have to set up your board for battery monitoring (option 3 : total voltage) by soldering the voltage dividers and connecting battery voltage to AN0, which is explained in the wiki here : Voltage sensors on APM1
To save processor cycles I've opted to drive the portK of the atmega1280 using direct portmanipulation.
If the above all sounds too complicated : here's the simple explanation.
In the usercode you'll often find something like PORTK = B00101010; This instruction actually turns leds on and off. 0 means off and 1 means on. Each number turns 1 port off or on, and the first number after B turns port AN15 on. So the ports are B-AN15-AN14-AN13-AN12-AN11-AN10-AN9-AN8 of shorter written : BAN15...AN8
If you would now like to turn the red leds on and the others off in the picture above, this would be PORTK = B00000001; (turns AN8 on, and AN8 is connect to nr1 on the darlington, which turns on red leds)
If you wanna turn the 2 blue ledstrips on and the others off it would be PORTK = B01010000; (ports AN12 and AN14 on, which are connect through the darlington to blue leds)
All 4 strips on would be PortK = B01010101;
Ofcourse for hexa's and octo's this depends on which AN port you have soldered to which nr on the ULN2803.
Now we can turn leds on and off, but they still don't blink or "rotate". For the blinking a counter is used and depending on the counter, leds are switched on and off. The counter runs in the 10hz or 50hz loop, so you can time the blinking. If counter =5 when in the 10hz loop will correspond to half a second.
For "rotating" arms I use the bitmath again, more precisely the bitshift operators << and >>. These simple shift all bit a few places. so 00000001 << 1 becomes 00000010. The 1 shifts 1 place. By now you understood that this is an easy way to make the leds "rotate". Just shift the B00000001 1 place each loop cycle and your 'on' leds will shift an arm each cycle, making them look asif they rotate.
4) TO DO
On the 8 port ULN2803 there are still 2 ports left unused on my hexa. I'm planning to use those to replicate the gps status light and maybe a failsafe warning light. I've also thought about using them to drive leds in the form of a smiley :-) Other ideas are always welcome !
I'm using the relay now to ignite fireworks on the hexa. But that's a story for another blog ;-)
That's about all I can think of telling you, and by now my fingers are tired and my brain is mushy.
I hope you enjoy these leds and don't scare too many people with your new UFO ;-) Feel free to ask questions in the comments below if something is not clear.
Ow, and ofcourse I assume that you know what you're doing if you solder on your board. I take NO responsability if something goes wrong and you break it ! Do this at your own risk !
Finally I've managed to have some time and do a little flight to test the Arducopter v2.1 firmware.
Here is my quad and some specs I hope you find them usefull somehow :
Weight (with camera on board) : 1700gr
Weight (without camera) : 1500gr
Battery : Turnigy 5000mAh 3S / 30-40C
Motors : 4 x EMAX GT 2812-11 930KV (about $15 each from plus2city - behaving very well so far)
ESCs : 4 x AE 30Amp (about $12 from hobbywing, behaving surprisinglly well so far)
APM 2560 + oilpan + Ardcuopter 2.1 firmware
12x4.5 propellers (not the best ones, this are a little soft, I'm waiting for some stiffer ones)
I got about 10m flight time with the camera
PID Rate Roll (and Pitch) adjusted to 0.075 (default was 0.145 but I got to much oscillations. The in flight PID adjustment using ch6 it's no doubt a super add on. I had never touched the PIDs before, but now I really needed since the quad was oscillating too much, especially on descents).
PID Stabilize Roll (and Pitch) adjusted to 4.550 (default was 4.5)
This 2 Videos, are from the same flight.
First one, videocamera from the ground (most of the time you don't see the quad, since I had no cameraman :) , but if you look closely towards the end, you will have a great chance to watch my belly eheheh)
Second video from camera on board. I left the time stamp so you can check the battery duration. Conditions were almost perfect. No wind at all, about 13ºC.
Posted by Ellison Chan on January 3, 2012 at 8:11pm
For those who haven't seen my previous blog here's some background on what this project involves.
The goal of this project is to prove that it is possible to print a miniQuad on a 3D printer at a reasonable price. This frame was designed using the excellent 3D modelling program Blender 3D. It would then be manufactured using a 3D printing service Shapeways.com. By using this printing service it becomes available for the public to order their own copies of this frame.
I am now one step closer to that goal. Over the holidays, I received the printed frame from Shapeways, and assembled it. Unfortunately, I need a few more parts, like additional motors, propellers, and batteries, before first flight.
My assessment so far is that, the strength and weight of the frame is more than adequate for flight. The plastic used has a small amount of flex, which should help to survive hard landings, but is stiff enough to keep the shape of the quad.
I have decided to dub this frame Firefly Mark One!
The following is major components for this project:
For those who always ask, "How much does it cost and where can I get one?":
I've placed all the parts of the Firefly Mark One frame in my shop on http://Shapeways.com/shops/rcshop, for public viewing. You can look at each model in 3D, spinning it around for better viewing.
The Firefly Mark One will come in at a cost of under $100, the long arm version of the frame (25.7 cm motor-motor), including the hub, 4 arms, battery holder, and APM mounting plate. Electronics, and mounting hardware not included, of course, since I can't print nuts and bolts.
Flight videos and tuning parameters coming, as soon as I get my parts from HobbyKing.
Would love to get any feedback from the community, before making it publicly available for ordering.
This definitely counts as more sci-fi than practical reality, but a South Korean team has proposed using a cloud of "robotic bees" to squirt building material to form structures the way bees make nests.
Writing in eVolo, an architecture magazine, Danielle Del Sol says, "These bees aren't interested in honey: these workers will actually build a structure. Each robot is capable of using cartridges filled with agents that enable them to construct literal physical material, which the designers dub "augmented synthetic material."
eVolo then offers this fanciful vision of a robotic bee:
Recently, I have been refreshing my knowledge of Machine Learning by taking Andrew Ng's excellent Stanford Machine Learning course online. The lecture module on Neural Networks ends with an intriging motivating video of the ALVINN autonomous car driving itself along normal roads at CMU in the mid 90s.
I was inspired by this video to see what I could build myself over the course of a weekend.
Posted by Wayne Dancer on January 3, 2012 at 7:42am
While looking at new products on GlobalSpec, I came across this sensor from Delta Metrics. Delta Metrics (a subsidiary of tecsis DBA) has created a single sensor to measure both static and dynamic pressure simultaneously, reducing the complexity of implementing an airspeed sensor. More information is available through Delta Metrics.
We often get questions from newcomers who are confused by all the Ardu* nomenclature here, so I made the above product comparison matrix, which you can find here.
And just to extend this to the full family, including software:
APM 1&2: the autopilot hardware (see above) that can operate nearly any vehicle, once the right software is loaded
ArduPlane, ArduCopter, ArduRover, ArduBoat, etc: the software that runs on the APM hardware that's optimized for that sort of vehicle.
Mission Planner: the desktop software that loads the software on the APM board, configures it and serves as a ground station, mission planner (of course!), an interface with flight simulators for hardware-in-the-loop simulation and a datalog analysis tool.
Posted by Mark Harrison on January 2, 2012 at 7:03pm
The 2012 AMA Expo (happening this week) will be featuring the NASA DROID, a drone that's being used as an educational vehicle for high school students.
Come and see and learn how NASA Dryden Flight Research Center uses the DROID or Dryden Remotely Operated Integrated Drone for flight research, pilot training and the INSPIRE student internship program.
The DROID is one of four aircraft used for INSPIRE or Interdisciplinary National Science Project Incorporating Research and Education Experience. INSPIRE is a multi-tiered year-round program designed for students in ninth-to-12th grades who are interested in science, technology, engineering, and mathematics education and careers.
“The INSPIRE summer internship program provides the opportunity for students interested in careers in engineering to get direct project experience prior to entering their senior year of high school or first semester of college,” said Candace Clements, Student & Faculty Programs Coordinator at NASA Dryden’s Office of Education.
Posted by Gareth Rens on January 2, 2012 at 2:00pm
Gyro stabilized mount works quite well. The roll servo does seem to be a bit jerky, but from the forums it seems its one of the downsides to digital servos. What about a small air shock absorber? A trade off of reaction time for fluidity of movement. Although, the platform is for photos not video...
Comments any1? :)
G:)
P.s. When am I going to be able to use the APM v2.0 for the above??? :)
A somber ceremony begins, as the machine which caused Major Marcy to sever all contact with us is dismantled & returned to its role as an aircraft position tracker.
The construction of an indoor test harness greatly improved matters.
A single steel wire from floor to ceiling arrested all horizontal movement & allowed us to focus just on altitude hold.
There you have the 1st successful altitude hold. This controlled altitude directly & was sloppy.
Camera mounted on the laser painting turret is tracking it.
The most successful algorithm controlled climb rate. The mane factor was update rate of the position. The camera could update position at 15Hz.
Focusing on stable takeoff & altitude with cyclic changes should keep our nights occupied. The real challenge is the horizontal position.
Now video of the vision guided takeoff & altitude control.
Just built a suprisingly good flying quad. Total price with battery but without receiver and APM was $127. It's very usable indoors and later I will add GPS mag, xbee and fpv tilt roll cam today.
I've put together a proof of concept autopilot based on the Lyapunov Vector Field path following stuff documented in various UAV research papers of the last many years. It is here:
It is written in javascript can can operate in one of two modes. It is only tested on Chrome.
a) it can fly JSBSim, using a local python connector that you run on your PC (but the web page is still remote). In this mode it uses a websocket to communicate with JSBSim. I use this mode the most as obviously the flight reactions are closer to reality.
b) it can run entirely in the webpage using a very un-aerodynamically correct flight model (I hesitate to even call it that). but at least one can still see how the navigation works without installing JSBSim. In this mode, there is no need for turn coordination, lift is very directly a function of elevator position, there is no stall possible, etc.
The dumb local world aircraft is a sort of Rascal-like UAV, and there is no point trying to select faster speeds or other planes even though there is a drop-down for that.
Wind can be simulated (in both modes). In the JSBSim mode, JSBSim generates the wind. The program modifies the wind speed and direction constantly if you want a "gusty" mode. In local mode, wind just moves the plane around as soon as it leaves the ground. On the ground, wind magically gets turned off lest the plane get pushed sideways etc etc.
It has other features
* A google earth plugin view from the simulated UAV. Press Down arrow in that sub-window to get a straight down ground view, left and right to look over each wing. Etc. If your PC is not fast enough, do not press the 3D button to enable the plugin. Move the plane to UAV to sydney, then plot a course and watch.
* Flight plan entry and editing from takeoff to Landing. Way points can be moved, deleted, or edited (using the right button on the marker). Given the performance of the plane, hit radius rings show goal satisfaction boundaries.
* Landing works in JSBSim, I can put in a pattern with a leg to final etc and if the altitudes are not unrealistic, it will hold wings level glide then flare. Takeoff is done by holding ground track with rudder and wings level till entered altitude AGL is attained.
* You can save and load flight plans (uses browser "localstorage")
* All the various autopilot and navigational goals and targets are viewable and you can drag the plan around in flight, or pause/start. You can also pause and move the plane then go again
* The vector field for any segment can be plotted, and there is also a mode where a click shows the surface from that point to the destination. (The surface is what the navigation autopilot tries to lock into by manipulation of the signal to the heading hold autopilot).
* The navigation autopilot has a critical parameter (Alpha) regarding the underlying heading hold autopilot lag. For fast responsive planes, this goes higher. For slow ponderous planes, this goes lower. At least in testing this number is always "ok" at 1.0, and once tuned for a particular UAV, insensitive to wind & speed variations. There is an easy way to find it, it could even be automated.
* During the simulated flight, when the UAV gets locked onto the surface, a green line shows the rest of that surface. You can vary how aggressive these vector fields go towards tracking.
* There are PID controllers "below" the navigation autopilot, however only the aileron, elevator and rudder are PI loops (the I is there to reveal trims when trims are incorrect or unknown), the rest are straight P (proportional control).
The arc following features, with approach, set entry and exit points, means it is possible to define how a UAV might execute a "corner" of a plan. Should it cut the corner? should it fly over the waypoint and do a wide turn? With a more advanced flight plan editor all this is possible if you match up the likely exit path from one track to the entry of the next. An arc can become a spiral as well: just vary the radius slowly as it is being flown.
The world data for both JSBSim and "local dumb physics" mode is 100hz. This is unrealistic with a GPS at 10hz. However the navigation level autopilot does not need a rate of 100hz, it needs accurate ground position and ground speed matched with current airframe orientation and motion. Therefore one could run the higher level loop at 10hz and the lower level heading/altitude/attitude/speed holds at 100hz, or one could dead reckon the GPS data and run at whatever rate.
I've done some testing with the math, the nice thing about this is that the config values are real world based: set speeds, roll rates, attitudes etc. The remarkable (at least to me) part was when I switched from the JSBSim Rascal, which is a small very powerful (for its size) UAV, to a Cessna 172, without changing any parameters except for V1 rotate speed, cruise speed and maximum desired roll rate, the flying and track following was still satisfactory out of the box, and became perfect when tuning the Alpha value to account for the slower nature of the Cessna control surfaces. Both would track a 200m radius circle in a prevailing or changing wind. All other settings were the same.
In the picture above the fake plane has tracked the curve accurately, then attained the surface to the next goal ("Surface Error: -1.3deg"). with a pronounced sideways attitude due to the wind from the right.
Also shown is that at 100hz, it needs 1800 trig operations per second, and about double that in flops.
Posted by Chris McNair on December 31, 2011 at 7:51pm
From all involved in the AttoPilot crew we would like to say Happy New Year and safe flying! Let 2012 be the year we show the Govt. who is the real boss!!!!
We wish you all the best in your new year adventures and hope that all are successful in everything they try to accomplish!
Keep up the great development work all and always think outside the BOX!
Keep your eyes open for new and innovative products from the AttoPilot, RP Flight systems and Hangar 18UAV this year.
We have some nice additions to the sUAS world coming to a store near you!
Put together a quick series of my favorite clips of MyGeekShow over this last year, and quickly discussed 2011's goal and how it was acomplished with the Nova.
Was able to go from building simple gliders to building the Nova - an airplane completely designed, built and flown by me, a huge success as I can now create an airplane to specifically meet the needs of any objective.
Thank you, thank you, thank you for all your support DIYDrones and members! Your support has truly made this show possible. Thank you!
As a hint for 2012, I'll be mastering FPV and AP... so you can expect future content to be more applicable to you in the coming year.
Posted by Mark Harrison on December 31, 2011 at 2:56am
Didn't intend to take any pics (the hat portion of ye olde hatcam being misplaced), but things went so well I was able to fly one-handed and hold the iphone in the other hand. Beautiful!!
From the air. I hadn't planned on doing this either, but things were so smooth I taped the GoPro and sent it off. Simple mode is going to be a Godsend for arial photography.
Spinning one-handedly. This is cut from the first video, but it's so unique and impressive I wanted to hilight it by itself.
