After noticing a friends SU-31 collecting dust in their basement, I asked to borrow the plane for a bit. Journey home, install a receive, download the manual,setup the control throws and expo - easy breasy. However, I did have trouble setting the CG just right and needed alot of weight on the tail which caused me to add the GOPRO which set it perfect - according to the manual.
Fast forward to the field - all setup and ready to go - give it some throttle - boom - blurry video (which I fail to clean). However, that's not the worst part. Final get to take off and this is pretty uneventful, but then I get in the air - super unstable. The CG is completely wrong and the control throws are wrong (luckily not terrible wrong). OK, well I set them to spec so maybe the battery got loose or I lost the camera on takeoff.. After a few minutes I get use to the way-off-CG/control throws and pull off a rather short landing - phew - awkward phone call averted.
So, I check the CG and throws again when it dawns on me, I'm looking at an E-flight manual, but this plane is from Great Planes!
All in all, a very lucky day for me and another lesson learned - and rather cheaply (one broken prop).
Compatible at 100% of the purchases ! We finally have some 6-Position mode switch DIY kit, compatibility info for new users !
Tested compatible Transmitters:
Turnigy 9x 7 Users
Turnigy 9x ER9x Firmware 4 Users
Futaba T10CG (removing the CamPac unit) 1 User
Futaba T6J (physical modification needed) 1 User
FlySky FS-TH9B 1 User
FlySky FS-TH9X 3 Users
Hitec Aurora 9 1 User
Additional user comments received:
> The switch works just like a dream. Thank you for a perfect solution !
> I tried out the switch and definately feel that it is worth my effort to install as now I can use 6 flight profiles and easily switch between them.
> I found the installation easier then soldering on a deans connector, and with my intermediate soldering skills I was able to complete the installation.
> Using this switch has been flawless. Once calibrated I have never experienced any failures with the hardware. It has come to the point that I trust it without a second thought. It has been so essential to my radio setup that I cannot imagine using an APM2 without it.
> Very Easy to set up. I am very pleased with the switch. And easy to mount when I can buy the parts in a kit and with instruction.
> installation was easy. i mounted mine in place of the hover pitch pot. i had to file the body of the "ele" switch a little to get it to fit, but it did -FlySky FS-TH9X
> It surely does make life easy... The big advantage is that now I have all the modes on just 1 switch. No need to look around for combination of 2 different pots...
> i am using it for my NAZA GPS and it made my life easier, one is used for flight mode (Manual/Attitude/GPS/Fail Safe(RTH)) and one for Intelligent Orientation Control (IOC) (Off/Home Lock/Course Lock)
> Extremely easy to install, even with it being the DIY version.It is perfect! I Could only use 3 modes easily before, so this is a night and day difference.
Harvard has been working on a robotic bee for five years now. Five years is a long time in the fast-paced world of robotics, but when you're trying to design a controllable flying robot that weighs less than one tenth of one gram from scratch, getting it to work properly is a process that often has to wait for technology to catch up to the concept.
The RoboBee has been able to take off under its own power for years, but roboticists have only just figured out how to get it to both take off and go where they want it to. Or at least, they're getting very, very close, and the latest testing was presented at one of the opening sessions of IROS this morning.
With the addition two small control actuators underneath the wings, RoboBee has been endowed with the ability to pitch and roll, which is two thirds of what it needs to be able to do to be a fully controllable robotic insect. These maneuvers are currently open-loop, which means that the RoboBee isn't getting any sensor feedback: it's just been instructed to steer itself in one particular way, which it obediently does until it violently crashes into something:
As you can see, these are hardy little robocritters: the prototype RoboBees have gone through dozens of flights, "almost always with crash landings," according to the researchers.
The reason that RoboBee hasn't yet learned to yaw is that all three axes of motion (yaw, pitch, and roll) are coupled together such that it's difficult to get a pure output with a pure input: if you try to get the robot to pitch, it's going to yaw and roll a little bit too, and isolating yaw from pitch and roll proved to be particularly tricky. Ongoing research will develop a feedback controller that can compensate for this, which should (we hope) mean that a RoboBee capable of hovering and fully controllable flight will be buzzing our way sometime soon.
"Open-loop roll, pitch and yaw torques for a robotic bee," by Benjamin M. Finio, and Robert J. Wood from the School of Engineering and Applied Sciences and the Wyss Institute for Biologically Inspired Engineering at Harvard University, was presented today at the 2012 IEEE/RSJ International Conference on Intelligent Robots and Systems in Vilamoura, Portugal.
It has been some time since my last blog post. I have kept myself busy with work and projects. For the last couple of years I have been working on trying to develop an airframe that combines aesthetics with functionality. Some may remember some of my previous blog posts where I presented my attempts this.
I have never been satisfied with most of the kits offered through other retailers. They all seem to incorporate the same parts and means of construction; lots of brackets, screws, and wiring that just decrease overall efficiency. Popular consensus is that carbon offers the best rigidity and overall structural integrity, but at a cost. I knew that there had to be a way to bring the cost down of a high quality carbon airframe without sacrificing functionality. At the same time I wanted to offer a higher degree of aesthetics than can currently be found without breaking the bank.
After many prototypes this is what I have come to. The construction is about as simple as it gets, and really light. I have greatly minimized the amount of hardware needed to assemble a solid carbon frame that can handle some of the hardest of tumbles. Many of my previous prototypes utilized a large number of machined parts, which greatly increased the total cost to produce. This version uses less than half the amount of previous versions. The bodies are thermoformed out of ABS or Kydex. The version seen here is Kydex. The bodies are easily trimmed and then fastened to the airframe using two small nylon thumb screws.
I have been a long time supporter of the APM and DIY Drones. You simply cannot get the robustness of an APM out of an off the shelf autopilot, especially at an APM price point. Yes, it requires tinkering and there is a learning curve, but it is well worth it when all is said and done. I remember a time not too long ago when the APM1280 was barely more than an expensive stabilization unit, and there were less than forty members in the Arducopter user group. Boy has the project come a long way, and it just keeps getting better and better. This is why all of my builds are designed around the APM. Next will be the PX4.
I wish I had some good flight video of the unit to post a link to, but this was a custom order AP rig. A few test flights, some tweaking and it was shipped off. In the meantime I am working on refining some of my fabrication processes to get the production time and cost down. The airframe pictured would cost less than half that of a comparable airframe If I didn't have to outsource some of the machining. My hopes are that after selling a few more of these units I plan to buy my own desktop milling machine where I can cut the parts myself. I already have the parts in Solidworks and Sketchup, just need a machine. My goal is to offer a high quality and aesthetically pleasing airframe at a fraction of the going price. Almost there.
Posted by Duran - ALTI on October 8, 2012 at 5:00am
Hey again everyone! Hope you are all, like us, having fun flying more than tuning, fixing :) Just thought we'de share some new images of our latest units, all using APM2/2.5. With our large heavy lift drones we've contemplated moving to ZeroUAV for the accuracy and 'easy to fly' ness of it, but I still LOVE the feel and ease of use with APM as well as the mission planner. We get asked many times why we don't use 'higher quality' flight controls on our rigs, but our experience in reliability and ease of use for our clients is super!
We've now probably done just shy of 10 000 flights with our own drones and test flights for clients without any issues, yes we've crashed a few time but due mainly to us being silly trying to flip hex or octos, fying from moving vehicles and other interesting things ;) 2.7.3 is pretty amazing and the code is getting better and better everyday. So Thanks again to all the devs and testers here for the hard work and great code.
Here's a ffew quick images of our SteadiDrone, QU4D, H6X and EI8HT Ultra, all running APMs.
A colleague of mine asked me to build a quad copter for him for where he can do photography with his GoPro, so I drew up some prices and with the 3DR quad frame and a GoPro Gimbal it was over $300 excluding motors and esc's so I went on ebay and found the frame above (Quad Frame on Ebay).
First I was concerned about those fiber glass motor mounts so I ordered replacement parts from small cnc parts to replace these motor mounts (Small Parts CNC on Ebay)
I had to drill one hole per arm to fit these but once these were on it was very firm no vibrations was going to get the better of these arms. I baught the 850kv motors and the 20amp esc from 3DR and those extended wires worked very well and neat going thru the quad arms.
I also baught the 3DR PDB and to my surpise it is quite a snug fit between the two main plates (one has to drill new holes to mount it though)
There is only two bad things from this frame
1. No parts available
2. Between the legs, battery mount and gimbal there is no neat way with the 3DR sonar mount to mount the sonar.
What nice is about the gimbal is that there are 3 servos 1 big one and 2 small once and on of the small ones is mounted in such a way that it can push the camera button, its got a nice arm and everything for this.
But I must warn you this is a bit of a lego/puzzel to build this quad you have to think a bit.
I then loaded the standard 3DR parmas for a base to start tuning the PID's (3DR Quad Parms) an to my surprise this quad ended up more stable than my own 3DR quad frame without any PID tuning, see video below.
After a rough start in Summer 2011, pan/tilt vision systems have emerged as the ultimate solution for indoor navigation. No surprise, after the success of the Kinect. This was the most advanced, so far.
It wasn't as stable as the AR Drone, but the advantages were size & absolute coordinates. It took a few years, but the budget & technology were finally in place to make an indoor quad copter capable of flying in the world's smallest apartment work. Now it's off to a customer.
The breakthrough with this system was conversion to 1 bit & run length encoding of the image on the microcontroller instead of JPEG compression. That allowed 320x240 to go at 70fps. 640x240 went at 40fps & turned out to be the optimum resolution. 40fps gives 20 unique position readouts, enough to maintain very tight altitude.
Since 1/2 the rows are skipped, the LED sometimes gets lost, but all the columns are scanned in 640x240. Ideally, this would use an FPGA & do 640x480 so every pixel would be covered.
The other factor making it possible was the arrival of copters stable & cheap enough to do the job. The Syma X1 is more stable than anything else tried. It has no accelerometer, yet automatically levels itself & resists horizontal motion. There seems to be a gyroscopic effect from the propellers.
It needed only a magnetometer for the autopilot to detect heading. It was level enough that a decent heading could be determined without any tilt information.
The Blade CX2 was also tried & found to be hopeless. That would not level itself, for some reason. There's hope a Blade MCX will be more stable, but the CX2 is still the only thing small enough to fit in the apartment, with a reasonable payload capacity.
The TCM8230MD, STM32F407, & RTL8192 combination has emerged as the ideal jellybean camera solution. That can do the high framerates, manual exposure, multiple camera synchronization & custom encoding you need for machine vision.
It turned out there was a blanking interval on the TCM8230MD where you could pause the camera clock & restart it to synchronize multiple cameras. It didn't affect the exposure.
The mane problem with the pan/tilt camera is determining where it's pointing. The direction derived from the servo PWM doesn't completely agree with the direction in the image. There's also wobble & delay in the servo motion. This creates a position which constantly drifts & has noise.
The ideal solution would be stationary markers in the room, which show up in the image & give the cameras an exact readout of where each frame is pointing. The most practical idea is 3 gyros directly on a camera for an instantaneous pointing direction which is blended with the PWM pointing direction.
There are ideas to improve the background separation. The flashing LED works really well, but alternating colors would work better. It's really time to start using FPGA's.
Posted by Swaroop Kagli on October 7, 2012 at 12:15pm
I came across OpenUPS. It is a UPS for small embedded systems with DC input and output (6-24V, ). It can manage batteries (any chemistry, LIPO? Of course!). It also has software interface for configuring it and control (need to explore this).
But the downside is - it can provide upwards of 6V only. Which mean we still have to use some linear/switching regulator to get the usual 5V the a BEC traditionally was employed for.
Otherwise, looks like a good intelligent compact power management unit. Very useful, I suppose, when we want to build bigger things :)
Came across this great product for antenna building for UAV's. Adhesive backed copper foil 5 mm wide. Great for creating your own dipoles, vee's or helix, easily soldered. Attach it to fiberglass or plexiglass or directly to the foam (prepped with acrylic spray to help adhesion or just taped over with package tape). Create dipoles or vee's directly on you wings (not over your carbon fiber spars though).
The non flying weather took a little break; this was a good time to test Mission flying with my T-Rex 450.
The sea breeze and the SW wind were fighting each other, so the flight was at times quite bumpy. As you can see on the video, the Heli was battling it very well, and the APM 2.0 with firmware 2.7.3 did a great job.....
lots of thanks to the DIY Heli team!!
The Mission was straight out, and then coming back in a zigzag route with sharp turns.
I reviewed the logs at home, and the Mission was very much on track.
I’m really happy with this achievement!!!
A small glitch I like to mention. When the Heli likes to land at the end of a Mission ( I know this is a bug) and I interrupt that and switch to Stabilize and then back to Loiter, the heli will drop to the ground. I experienced it two times. After going to AltHold and then to Loiter, everything works. Looks like the landing information have not been reset, I don’t really know.
Ours started small, almost unnoticeable, but gradually got worse till it was very annoying.
Now we can identify the GoPro buzz in many videos folks post on the net.
We chased our small buzz in our video for days. The buzz continued to get worse as we kept attempting to isolate the annoying sound. If we shook the Go Pro a certain way, we could hear a very faint little rattle. This was the little rattle that was amplified in the GoPro case. We found for us the little plastic lens on the front of the Go Pro was the culprit. The lens is held in place by a two small little round protrusions built into the front half of the case. When the lens's are installed, it appears the little protrusions are melted to hold the lens in place.
Well, ours worked its way loose. The solution for us was to put two very small dabs of epoxy next to where the little protrusions are, holding the lens tight to the case. In the picture, GoPro Buzz you can see the two small little dabs of shiny clear epoxy.
Hope this helps and saves some time for someone :)
There are alternative antenna choices for your UAV when space is a consideration. RHCP antennas are without doubt one of the best choices you can make for your video and telemetry signals. I had wanted to use a RHCP antenna for telemetry for my X-8 but the choices I had come across were pretty space demanding. Surfing the web doing some antenna research I ran across these wheel antennas from Kent Britain WA5VJB, a fellow amateur radio operator. www.wa5vjb.com He's got some really useful designs other than the wheel. I purchased 2 of the 900 mhz variety for the telemetry, one for in the X8 and one for the ground station . Since the plan is to install all antennas within the X8 these are a great space saving choice over the current offering of cloverleaf antennas. The 900 mhz wheels measure about 4.5 inches in diameter, somewhat smaller than a CD ROM and only slightly thicker. The best part being they are only $ 12.00 each. They can be fed with coax directly or as seen in the photograph you can solder a circuit board SMA connector of your choice to the center in the holes provided. These wheel antennas are considered to be horizontally polarized at the plane of the horizon however as your UAV gains altitude the wheel antenna emits a cross polarized signal. Here is a quote from hamtv.com “Around 1990 the first launch of a balloon borne payload in the Denver, Colorado Area took place sporting a Little Wheel as the radiator for the on-board ATV transmitter. It was found to work quite well and had some interesting attributes that were previously undisclosed. Those being that the Wheel antenna although being a horizontal radiator on the horizon displayed properties of cross polarization above and below the array - similar to two in phase horizontal dipoles. This worked to the advantage of the ground station by employing a circularly polarized antenna (Helix) - also a fixed horizontal antenna directly below at the launch site would not see the up to -30 dB nulls as the balloon rotated. The horizontally polarized tracking antennas (Yagis) were receiving a signal of constantly changing polarity as the payload rose to an altitude at which it would start to drift away from the launch site and the horizontally polarized tracking antennas would become more effective due to the horizontal polarization emitted from the sides of the Wheel. The pattern of the Wheel is within +/-0.5dB of being a perfect circle in the E-plane. In the H-plane the pattern is +/-35 degrees of the horizontal axis of the antenna. These figures equate to ~ 4.28 dBic gain or 6.38 dBdc. The dBic gain above and below the array is somewhat indeterminable due to the fact that the circular polarization exhibited is quasi-circular or cross polarized; however, it responds to a Helix in a manner consistent with the gain of a pair of like sense Helices in a system looking at one another. ”
This antenna does radiate well (60% efficient) and also radiates well in the second and third harmonics so take precautions to provide filtering for your transmitter. You’ll want to use a “Wheel” antenna on each side of your link as losses will result if mated to just a dipole when your UAV is overhead. Just mount it in a horizontal position at each end of the link. If it fits inside your UAV that’s great but it could also be mounted flat on the wing as well.
Holes are provided for SMA circuit board jack mount or direct connection to coax.
This link shows the antenna being tested in a 3D pattern.
Sidney Leggitt has launched a project on Indiegogo to start a company to create a "Personal Drone Network". At the moment, that appears to be mostly just selling a RTF ArduPlane for $850 or $1,500 with a full video setup. Looks like a pretty cool project, but it would have been nice to credit us, given that he is using our technology and copied the marketing language (including my mistakes!) from some early APM product listings.
I just noticed on Popular Science that Discovery Channel is putting out a new documentary called "Winged Planet". Looking at the previews, I was delighted to see that they have built a FPV Vulture with HiDef cameras, and they also used a multicopter for other shots.
We describe our autonomous vision-based quadrotor MAV system which maps and explores unknown environments. All algorithms necessary for autonomous mapping and exploration run on-board the MAV. Using a front-looking stereo camera as the main exteroceptive sensor, our quadrotor achieves these capabilities with both the Vector Field Histogram+ (VFH+) algorithm for local navigation, and the frontier-based exploration algorithm. In addition, we implement the Bug algorithm for autonomous wall-following which could optionally be selected as the substitute exploration algorithm in sparse environments where the frontier-based exploration under-performs.
We incrementally build a 3D global occupancy map on-board the MAV. The map is used by the VFH+ and frontier-based exploration in dense environments, and the Bug algorithm for wall-following in sparse environments. During the exploration phase, images from the front-looking camera are transmitted over Wi-Fi to the ground station. These images are input to a large-scale visual SLAM process running off-board on the ground station. SLAM is carried out with pose-graph optimization and loop closure detection using a vocabulary tree. We improve the robustness of the pose estimation by fusing optical flow and visual odometry. Optical flow data is provided by a customized downward-looking camera integrated with a microcontroller while visual odometry measurements are derived from the front-looking stereo camera. We verify our approaches with experimental results.
Here's the team and their birds. Huge and well-deserved congratulation!
Posted by Peter Spero on October 5, 2012 at 7:30am
Hi All, my name is Peter Spero, I am an musician and composer from Chicago. I would like to create original musical scores for pilots who are capturing media from their UAVs. You can check out some of my music at http://www.SperoMusic.com. Much of the music on this site are my POP creations but I can also go way to the HipHop and Electronica side. Thanks All!!! There's no cost involved for now so please let me know if you are interested.
Posted by Stephen Dade on October 5, 2012 at 4:30am
We won't be able to begin the (long) process of uploading the photos and imagery of CanberraUAV until we get back home, but I've uploaded a few of the "best of" photos taken from our UAV inflight:
