Well, quite a lot has happened since my last post on here. The biggest news is that I'm in Peru! In the 2 1/2 months since my last post, we built and tested all the aircraft form the previous post and have gotten them in good working order. We also took delivery of a FLIR A300 thermal camera along with the requisite wifi systems to control it while it dangles under the hexa. Before I move on to details, I'll share a picture of my current base camp in the Qorichaska Hostel in Cusco.
Base camp. In the frame are our 2 gigantic Pelican cases, telemetry tripod, toolkits, computer, and my clothes. Numerous other bits are scattered around the room.
The new aircraft have not expanded our capabilities all that much, but they have made our lives quite a bit easier. The only significant expansion of capability is with the new plane, which does seem to provide the 20% increase in flight time advertised by the manufacturer. This is the result of the wing re-design which, if my feeble non-engineer's understanding is correct, have a longer chord and shorter length which has decreased the amount of drag produced at cruise.
Other than that, the shorter wings have made the plane significantly easier to pack and the benefits of the hexa's ability to fold should be clear enough. Switching from an octo to a hexa also decreases complexity and makes construction, maintenance and upgrades 25% easier.
Well, sparing a lot of details about the build and test process of the new aircraft, that brings everything up to speed. My field assistant, Billy, gets here tomorrow morning and then on Monday we are off to Villa Carmen, a field station east of here owned by the Amazon Conservation Association. It is at only 520m of elevation vs the 3800m of Cusco, so it will be a good place to make sure everything still works properly before we stress the aircraft with extreme elevation and steep terrain. Fortunately, Villa Carmen does have a ridge at 1100m within a few km of the station, so we will be able to test performance on the slope.
So, I'll give another update or two from Villa Carmen about how things go there and the plan for the 3 weeks of flights in the mountains that will come after.
We have been lucky to receive funding recently from the Sabin Family Foundation, along with our long-time supporting group, the WFU Center for Energy, Environment, and Sustainability. With this critical funding we have been able to take our operation to the next level and get the equipment we need to carry out our planned missions in Peru. Here is the short list of what we have gotten recently:
2 new 2014 1800mm Skywalker airframes, one of which will be built out into a primary aircraft, the second of which will be maintained as a spare
2 new Tarot t810 hexacopter airframes. As with the fixed-wing airframes, one will be built into a primary aircraft and the other will be a spare
A Honda eu1000i 1000W generator to support sustained field operation
A GoPro Hero3+ Black edition to improve the quality of our aerial video
Assorted FPV equipment to increase the range, reliability, and flexibility of our FPV video links
An additional Sony NEX-5 camera to provide redundancy in case the primary camera fails
A 16mm prime lens for the NEX to allow it to be used on the DAO
A Canon SX260HS modified to collect imagery in the blue, green, and NIR bands, allowing us to collect blue-NIR NDVI (very comparable to red-NIR NDVI)
Here are a few pictures of the new gear:
Wide view of the new hexa. It is a beautiful machine.
The basics on the center hub. APM, Dragonlink rx, RFD900, HK F-40A ESCs, trays for dual batteries.
Downward-facing motors which should produce in excess of 2.5kg of thrust at WOT.
And the arms fold up nicely
As do the legs
Same Skywalker 2014 fuselage and tail with new 180 cm wings with downward cranked tips.
GoPro mounted on the nose for FPV. This is a Hero 3 White Ed but it will be replaced with a 3+ Black Ed as soon as the plane is through its testing phase.
Wings have higher chord than the old ones, which should provide more low-speed lift and efficiency.
A nice look at the broad, flat wings with cranked down tips.
I have built the new fixed-wing aircraft, we have begun testing it, and we are very impressed. Other than the crash, detailed in the earlier crash post, it flies extremely well, better than all other previous airframes. It is very responsive and controllable, probably due to the new wing design (the fuselage and tail are the same as our previous version). The ailerons are much larger which, while probably contributing to our crash, also makes the plane's roll performance much, much better than the old airframes (rolling the old DAO with aileron was like trying to turn a bolt with a ratchet set in the wrong direction). We will continue to evaluate the new airframe, including the manufacturer's claim that the new wings are significantly more efficient than the old wings. The claim seems very reasonable, as the wings have a much higher chord, making them perform better at low speed. If this is true it should put us well into the 1 hour flight time range.
I am currently in the process of building the new hexa. It is fully finished save for the propellers (we are still waiting on them to arrive). It is truly a beast, with T-motor MT2826-6 motors (producing peak thrust in excess of 2.5kg with 15" carbon fiber props). I have been very impressed with the frame itself, as well. It is very rigid, well put together, and folds up nicely while locking open securely. The only bad thing it that it was an absolute nightmare to assemble. As soon as we get some props for it I'll get it in the air and see how it flies. I expect to to be a wonderfully-terrifying mix of silky smoothness and raw power. It will need both if it's going to fly the thermal camera we recently submitted a grant for.
So, that gets the blog just about up to speed with what we've been doing lately. The only other update I need to give is on our mapping mission of Pilot Mountain. Look forward to that in an upcoming post.
Ok, so I couldn't help it, I had to put that title. Anyway, I have tons of great pictures from Belize, so I'm going to try to let the do the talking. To get things started, here is a picture of our home for 8 days while there:
It could have been worse...
I'm not a very good photographer, it's just the most photogenic place on earth.
I tagged along with the coral reef ecology class to Lighthouse Reef Atoll. The atoll is about 75 km off the coast of mainland Belize and is home to some of the best coral reefs and stands of mangrove in the western hemisphere. I went to finish what I started last year: map all the islands. I came armed with the DAO, all of the long-range radio gear, lots of spare parts, and Matt and Nella.
Our plan was simple: fly from Northern Caye to map it and Sanbore Caye on one day, then fly from long Caye to map it and Half Moon Caye on another day. We made the Northern flights first.
We flew from an abandoned 2000' airstrip there and made 4 flights. We flew half of Northern in the first flight. The second flight was aborted due to overheating issues. I had moved all the electronics inside the fuselage to improve resistance to a possible water landing but I had not forseen the heat issues. I moved the ESC back outside the fuselage in the field and got back in the air to fly the other half of Northern. After completing the imaging of Northern we turned our attention to the 3km flight out to Sanbore and back. This flight was over open water and was, by all accounts, risky. Regardless, we made it happen. All went well with the flight but, unfortunately, the battery in the imaging camera died just as the plane made it to Sanbore. We now have 4 batteries for that camera... In all, all flights from Northern were pretty uneventful and we got an incredible mosaic of the caye.
Preparing to fly
Under the hood
A couple days later we flew from Long Caye. Flying here was much, much more difficult. Instead of a 2000' airstrip and no wind, we had a pier and a 20x10m beach and 25 km/h sustained winds. We elected to take off from the pier as it was oriented directly into the wind and provided a good open departure path. I elected to land on the beach, as the 20m side was oriented parallel to the wind, making it the biggest open spot on the island which matched up with the wind. We made 3 flights here, 2 over Long and 1 out to Half Moon. All the flights were entirely uneventful save for a few telemetry and RC connectivity issues at the extreme ranges of our flights (~6km). Landing was the only difficult part. As discussed in my crash post, the second landing did not go well. The first was hard, as I came in fairly slow and when I descended below the treetops I lost the headwind, causing the plane to stall about 1m above the beach. No damage was done. The second flight ended in a minor crash discussed int he other post. The third landing was absolutely flawless but, unfortunately, the RC system lost link when the plane got on the ground causing it to failsafe and accelerate into a tree. Damage was minimal from that.
Despite good time in air, the imagery from Long was not all that good. We identified a major flaw in our imaging system. Our camera's CHDK script causes it to take a picture every 2 seconds while the camera gets a signal over USB and, when that signal is cut, it powers off the camera. Once the camera powers off there is no way to restart it in the air. We knew this but what we did not know is that our RC failsafes were set up to turn off the RC channel controlling the camera. This meant that when we got our first intermittent failsafe on our second flight over Long and our flight to Half Moon, the camera cut off. We ended up having excellent imagery of about 2/3 of Long Caye and none of Half Moon Caye because of this. We now have a 16mm prime lens for the Sony NEX-5 and will be using it in the plane now. Because the NEX uses an IR trigger this will no longer be a problem; we will be able to toggle between capturing images and not while in flight.
Ready to fly at Long
Runway, Long Caye style
Under the hood at Long
Long Caye GCS
In all, it was a great trip that got us a lot of great data and taught us a lot. We made some mistakes that now we don't ever have to make again (it is just unfortunate we made them there...). Anyways, check out the links below for some data products and more.
In this post I'll talk about the most important work we've done in the history of our (still young) program. In short, we provided imagery of an environmental disaster, evaluated its severity, and got national news recognition for it. I won't go into much detail and, instead, point you in the direction of some excellent articles about our work.
On Febraury 2, 2014, a settling pond for coal ash located alongside the Dan River in Eden, NC ruptured, releasing a massive amount of the toxic sludge into the river. Coal ash is a who's who of toxic heavy metals and other such nastiness and is commonly stored in ponds on the grounds of coal-fired power plants. Under the pond at the power plant in Eden there was a stormwater drainage pipe which, unsurprisingly given that it had never been inspected in the past 50 years, ruptured. This rupture allowed the contents of the primary ash pond at the plant to flow into the pipe and directly into the Dan River. In the days following the spill, Duke Energy provided estimates of the spill's size which were constantly in flux. We identified the situation as one that could benefit both from detailed visualizations of the spill site and a 3-dimensional model, which we believed could be used to estimate the volume of spilled material.
We were notified of the spill on February 6 and the next day we traveled to the spill site and flew the DAO over the ash pond. We collected imagery over the site at 90 and 115m AGL, providing excellent coverage of the area for good reconstruction quality.
Flight path of the DAO over the ash pond. The pond is not visible here but is, as you might expect, under the grid flight pattern.
We had flawless collection, processed the imagery in a few days, and produced visualization of the spill as well as what we feel is an excellent estimate of the volume of the spilled ash and toxic water. From there things went wild, WFU news put out a press release and the next day I did 7 interviews. I am still doing interviews from time to time and this mission will go down as one that brought enormous recognition to our lab, our university, and, most importantly, to the spill itself.
This also showcased the impressive capabilities of our aircraft to rapidly respond to a need, image an area, and produce valuable data products.
I’ll start off with the bad news. I’ll go over all of our
crashes since the last post. There haven’t been many and most were fairly
minor, so it isn’t all that bad.
Our first crash was with the DAO when we were attempting an
auto-takeoff. We had some false starts and then, with a running start, the
plane took to the air. Instead of following a straight-out heading, the plane
immediately banked right and made 2 climbing circles at full throttle before
nosing down from a relatively low altitude and hitting the deck. I have not
dissected the cause of this crash but we are pretty certain that we needed to
command APM to restart the mission after the aborted takeoffs before the successful
one. We think the plane was attempting to loiter at home when something caused
it to nose down. From this we learned not to use a front-line workhorse
aircraft for testing. Fortunately, damage was pretty minor and some glue got us
back in the air.
The second crash was when we went to Belize. This was all
pilot error. I was coming in for a landing on a very small beach under FPV
control. The beach was only about 15m long and 5-10m wide, and I had to make a
low approach over mangroves and a small lagoon to keep the plane oriented into
the wind. The mangrove canopy about 50m short of the beach necessitated a
fairly steep glide-slope. I was on my second approach to the beach (we had
already flown a portion of the island earlier) and from my view with FPV it
looked like I was approaching too low and was in danger of landing short in the
small lagoon. According to the spotter on the beach I was not, in fact, too low
but I was flying remotely so that was not relayed to me in real time. I pulled
up and added throttle to clear, which put me too high and in danger of landing
long and hitting palm trees at the edge of the beach. To avoid that, I nosed
back down hard, digging the nose into the sand and flipping the plane onto its
back. Damage was minor and repaired within an hour, allowing for a third flight
and a successful landing on the same beach. From this crash, we learned that
sometimes there is no replacement for an LOS landing.
Making repairs on the DAO after our hard landing on the beach
Our next crash came a little over a week ago. Marcus and I were working on ironing out issues with the DAO which had persisted since the Belize trip and crash. Repairs on the tail had left the geometry of it a few degrees out of alignment and something (probably exposure to heat during transit) had caused the wings to warp (a detail we didn't discover until after this crash). Due to these 2 problems, the plane had flown really poorly since returning from Belize. In the process of doing test flights we simply got too low got the black plane below the dark tree background, and lost orientation. The plane went down causing minor but not insignificant damage. From this crash we learned that the foam airframes are disposable and are easier to replace than repair. New airframes after significant crashes are just a part of doing business. As a result we retired the original DAO.
Our fourth and final recent crash was just a few days ago. We had gotten in our new airframes (which are awesome, but I'll talk about that in more detail in an upcoming post). We put the new wings on our new primary airframe to evaluate their performance. The new wings have a few significant differences from the old, including much larger ailerons (about twice the size as the old DAO wings). We aren't entirely sure what caused the crash during our test flight, as we shut down the laptop shortly before it happened (we were winding down for the day), but I am pretty confident it was a tip stall. We had already done about 15 minutes of flawless test flying and Tim and I were both very, very impressed with the new wings. We entered a slow left bank in FBW-A at about 40m AGL, with roll limited at 30 degrees. In the bank the plane suddenly rolled past the 30 degree limit into a dive and was unrecoverable. My belief, based on review of the onboard camera footage and the dataflash log of the RC I/O data, is that in the bank there was a gust of wind coming from the right of the aircraft, which caused the APM to drop the left aileron to fight the roll caused by the wind gust. This drop of the aileron on the slow wing (the wing to the inside of the turn) put the angle of attack too high, leading to a tip stall in the left wing. This caused the plane to roll over and dive. At this point the plane should have been easily recoverable, but I believe one or both of the batteries slid to the nose of the aircraft, moving the CG too far forward to be able to pull out of the dive. Fortunately damage was minimal and limited only to some deformity of the plane's nose. From this we learned that tip stalls are real and we need to keep airspeed up to avoid them. We also need to secure batteries better so an event like this doesn't lead to an unrecoverable dive.
Skip ahead to 10 minutes to see the crash.
So, that is all I have for crashes. It has been a pretty tame few months without too much excitement. Most of our flying has been uneventful and some it has produced downright astounding results... but more on that later.
Well, it has been quite a while since we have
posted on here. It hasn’t been the result of having nothing to report;
actually, it has been quite the opposite. Far too much has happened in the past
few months for one post, so the next few posts will be a recap of what has
happened. As those posts are coming out, I’ll give you a little overview. We
imaged an environmental disaster and made national news, we flew in Belize, we
acquired multiple new aircraft, and we crashed a few times here and there. I’ll
talk about each of those items in a separate post, so stay tuned.
Disclaimer: This post is as much for my own benefit as anyone else's. I needed to write down all the things that are swirling around in my head after AGU and I thought it might be interesting enough to others to put up here.
First, AGU went really well. Here you can see the poster I presented.
I had the great opportunity to network with a lot of people working in the UAV field and there were quite a few people who had a great deal of interest in what we are doing. I was able to meed John Dandois of Ecosynth and the University of Maryland (ecosynth.org), one of the true pioneers of using UAS technology in ecology. They are doing really great work on the same sorts of things as us and I definitely see the possibility of some collaboration in the future. I also met people from University of Arizona, Oregon State University, University of Alaska-Fairbanks, Norway, and San Francisco State University who are all working with UAV and ultra high resolution imagery. These connections all have the possibility for learning quite a bit from the mistakes of others and for them to learn from ours. Additionally, I had the chance to meet with some vendors, which is part of what I wanted to cover here.
I spoke with guys from Google Earth Engine and Microsoft Research. Earth Engine is really cool in that you can run some pretty advanced data analysis in the cloud, which would take a lot of the load off of us when generating really complex products with vast amounts of data. Then, with the ability to tie Earth Engine into Maps Engine, where we could upload imagery and have it processed by Google into more manageable pieces, the prospects for a really powerful data management, visualization, analysis, and dissemination tool start to become clear. This is something that I will be working with over the next few days and I am pretty excited about the prospects. Hopefully a grant from Google which will provide us access to these tools will really open up our possibilities in the data visualization area.
Microsoft Research has an interesting GIS program called Layerscape which looks to be much more user-friendly than something like ArcGIS and, as you might expect, it comes with all the necessary bits and pieces to disseminate the maps once you have made them. It looked pretty cool but it remains to be seen whether it will be able to beat the Google offerings and be worthwhile.
I also spoke with a couple hyperspectral imager vendors and people who are operating hyperspectral units on UAS, which got me thinking about these kinds of things again. I found a couple companies which have really impressive offerings, namely Resonon (resonon.com). I heard from a researcher I was talking to that the Resonon systems are actually fairly inexpensive, which would be huge for us if true.
Well, thats all for now, hopefully I have even more exciting stuff to share next time.
