FieldNotes

Stormy Weather

Anonymous — Fri, 08 Apr 2016 02:10:13 +0000

Stormy Weather Anonymous (not verified) Thu, 04/07/2016 - 20:10

Don’t know whhhhhy there’s no plane up in the sky, stormy weather, wind’s been staying high, since forever

Ok so I’m not exactly a songwriting genius.

The AMF is the small complex of containerss with the radars on the roof. This is our base of operations for the two-week campaign.

Conditions have been windy here on the north slope the past couple days. Gusts hit 51mph at the NOC camp yesterday and sustained winds offshore remained in the upper 40s. Too much wind to launch and land the Pilatus or fly the Data Hawks.

The high winds bring rapidly drifting snow. Every morning we have to wait while the road to the site is plowed out. Even then we have to return to the NOC camp in the afternoon before drifts build up again.

A loader clears the way to "Jim's Jail", a warm storage unit near the hangar where we keep and work on our planes.

Nonetheless, work continues, Gijs and Phillip built a rig that allows them to run the POPS on the roof of the AMF. So, even though we can’t fly the Pilatus, we’re still getting data. Meanwhile, several tests were run yesterday to help pin down the autopilot communication problem on the Data Hawks and Doug fixed the issue with the loose elevons.

Gijs and Phillip prepare a rig to allow running Pilatus sensors on the roof of the AMF while we wait for winds to calm enough to actually fly the Pilatus.

This morning winds calmed briefly and we made another attempt at making Data Hawk flights before they increased. On our first attempt the communications issues with the autopilot stimied all attempts to get it running before launch and we ended up doing a short flight on RC to see how the plane handled the conditions (winds were lower, not low, still blowing 25-30mph) and collect some data with the cold wire sensor. The flight was a short one, lasting only around 10 minutes but it was our first successful flight of the deployment. We had a minor mishap at landing, when a sudden gust knocked the plane into the runway and broke the prop, but it was nice to finally get a measurement flight completed.

Flying UAS in the arctic step 1: Drill a hole for your bungee launcher stake since pounding the stake into the frozen ground is impossible.

After recovering the plane, Doug replaced the broken prop and, with help from our colleagues in ��«��Ƶ, was able to fix the autopilot communication issue and we made another attempt to get an autopilot controlled flight in. Unfortunately an elevon came loose again and the flight had to be aborted. With winds building and snow drifting in again, we returned to the NOC camp for the afternoon.

Doug flies a Data Hawk under manual control while Phillip acts as observer.

Off

Traditional

White

ERASMUS Spring Campaign Start Up

Anonymous — Tue, 05 Apr 2016 06:00:00 +0000

ERASMUS Spring Campaign Start Up Anonymous (not verified) Tue, 04/05/2016 - 00:00

A small team from RECUV (Gijs DeBoer, Doug Weibel, Tevis Nichols and Phillip D’amore) arrived in the arctic on April 2nd for a two-week campaign for the Evaluation of Routine Atmospheric Sounding Measurements using Unmanned Systems (ERASMUS) project. The weekend was spent preparing aircraft and sensor systems for flight. All equipment for the two week campaign had to be shipped through Prudhoe Bay and out to the Atmospheric Measurement Facility (AMF) at the Oliktok Point Long Range Radar Site (LRRS). With no ability to order parts or tools at the site, absolutely everything we might need had to be shipped in advance. The monumental packing effort was executed flawlessly by our colleaugues back in ��«��Ƶ and as a result flight preparations preceded without any major issues.

Three aircraft were shipped to the arctic. Two small DataHawk aircraft, equipped for basic atmospheric measurements, and one larger Pilatus aircraft configured to carry radiometers and the Printed Optical Particle Sensor (POPS). With the full sensor load this aircraft weighs over 50lbs.

The Pilatus UA outfitted with RS-92 Pressure, Temperature, Humidity sensor, SPN-1 radiometers, POPS and CGR4.

Unfortunately, while preparations proceeded smoothly, the weather turned rough. Winds were moderate when the team arrived but began to increase into the 35-45mph range Sunday night. By Monday morning, a winter weather advisory was in effect, visibility was limited, and large snow drifts began to impede mobility on the roads. In such conditions, launching and landing the Pilatus is too risky to attempt and the slower Data Hawks are in danger of being blown away. As a result flight operations were postponed.

Conditions have been windy and a bit cold to say the least.

Instead, we spent Monday morning running through ground tests to prepare for flight operations when the weather eased. Monday afternoon was a bit of a break, winds remained too high to fly and with site access limited by rapidly drifting snow, we did what work we could at the ENI camp.

Tuesday morning winds eased into the low 20mph range and temperatures climbed above 0F; the decision was made to attempt a DataHawk flight. Unfortunately while the weather was warmer that’s a relative term and one of the DataHawk’s elevon servos (this aircraft has no horizontal tail so elevator and aileron functions are performed by the same surfaces) was responding sluggishly in the cold. In the end, the control delay, combined with the high winds made controlled flight all but impossible. Doug managed to keep the plane aloft after the bungee launch but shortly after beginning the turn toward the planned flight area, it crashed on the Tundra.

Gijs DeBoer and Doug Weibel prepare to launch a Data Hawk unmanned aircraft in challenging conditions.

Recovering a plane in the arctic is bit tougher than picking up a Tempest out on the plains in Colorado. The short walk to the crash site required three people, survival suits, and a shotgun to protect against potential polar bear encounters. Damage however was light, and with quick repair to the left elevon, and a new servo to replace the one which had responded slowly earlier, the plane was again ready for operations.

Unfortunately a serious problem with the autopilot was found during the attempt. All radio communications were lost between plane and ground station at takeoff and could not be restored. The brief flight was conducted entirely on manual control, but a solution to this problem is needed to allow for automated DataHawk flights in future. With winds increasing again at midday, we returned to the camp to work on a solution in the afternoon.

Off

Traditional

White

That's All, Folks!

Anonymous — Wed, 19 Aug 2015 23:37:24 +0000

That's All, Folks! Anonymous (not verified) Wed, 08/19/2015 - 17:37

With a final automated flight in the fog and a series of low level box patterns over the beach on Saturday, we concluded the summer ERASMUS deployment to Oliktok Point, Alaska. With over 150 aircraft launches, 50+ profiling flights, including 23 individual profiles in a single 12 hour block, 23 hours of flight time, and nearly 350 individual ground station starts, it was a very busy and flight intensive campaign. The vast quantities of data collected will go on to further knowledge of weather patterns in the arctic and how they effect weather all over the Earth. The experiences with the aircraft itself will drive improvements of both the airframe and flight controller, further increasing the usefulness of the DataHawk2 as a research platform.

The past two weeks made up a great deployment. From solving engineering problems to collecting valuable atmospheric data, this campaign has embodied what RECUV does best: using engineering strategies to design and integrate unmanned vehicle systems to accomplish relevant research in new and innovative ways. It's been great working with people not only from the CU Aerospace Sciences department, but also from CIRES, ARM, Fairweather, ARCTEC Alaska, The Pacific Northwest National Laboratory, as well as Sandia National Laboratories. Thank you to the Oliktok LRRS station technician Lynn Nelson, station chief Carter Boyd and the United States Air Force for making our stay comfortable and productive. Also a big thank you to Wessley King, the AMF-3 operator, for providing workspace and transportation as well as his time for our use. Finally, thank you to Al Bendure who provided assistance in innumerable forms including coordinating with the FAA, ConocoPhillips, BP, ENI, PNNL, and SNL as well as spending many cold, windy, and foggy hours perched in the back of the pickup truck on bear lookout. This deployment couldn't have happended without them, and to them we are grateful.

This concludes the field notes for the summer ERASMUS deployment. Expect further ERASMUS updates as the Spring 2016 deployment draws nearer.

Nate pilots the DataHawk low over the beach

Gijs and Al set up the ground station antenna

After the final flight! Thanks for tuning in! Spring 2016 is coming up!

Off

Traditional

White

"Now we're all muxed up!"

Anonymous — Tue, 11 Aug 2015 07:00:52 +0000

"Now we're all muxed up!" Anonymous (not verified) Tue, 08/11/2015 - 01:00

Multiplexered up that is!

Thanks to the heroic efforts of our ��«��Ƶ colleagues (being at DIA at 6:45am to drop off a package is heroic) and the amazing machine that is commercial air travel, the multiplexers arrived in Deadhorse on Saturday. Due to soggy roads and oil rig movements (the road is only just wide enough for them alone), we were unable to have anyone drive the 40 miles (64km) to pick them up until Sunday afternoon when our friendly AMF-3 operator Wessley King made a run to town. After an evening "build party", the retrofitted DataHawk2s were ready for flight. Unfortunately, 12m/s (27mph) winds with rain and 100m (328ft) cloud ceilings severely limited our flight opportunities for Sunday evening and most of Monday.

Finally, after scrubbing many of the scheduled Monday profiles, the clouds lifted up to 600m (2000ft) just long enough to fly 8:00pm and 8:30pm profiles. With the autopilot set to climb at 4m/s (13ft/s) up to 150m (500ft), both launches were done under automatic control, without even the slightest radar interference. From there, the aircraft profiled up to 300m (984ft), and back down at a nice constant 1m/s (3.2ft/s) several times. Everything was going as planned until the interference struck again and rendered our automatic flight controls useless. A flick of a switch was all it took to restore control to the RC transmitter and the aircraft landed safely back at the groundstation. While this was a major success for the multiplexers, the loss of autopilot at higher altitudes will continue to limit the profiles to lower altitudes than originally planned. As much as this is about not losing aircraft, it is even more about safety; even in restricted airspace on the North Slope of Alaska there are plenty of people and lots of buildings to which a fast-moving 1Kg (2.2lb) UAV would cause significant injury or damage.

Tomorrow, the weather is forecast to improve significantly, so we have 12 profiles scheduled hourly from 8am to 9pm. This is still dependent on the weather cooperating, but with the autopilot doing the flying and the safety of the multiplexer override, these upcoming flights should prove to be far less stressful.

We've made good progress in moving past the problems that have arisen during this deployment and with less than a week remaining until we head home, we're looking forward to collecting more data for a strong finish.

From the plane to the groundstation to bears, there's a lot to monitor while on deployment at Oliktok!

Nate finds the best way to pilot high altitude UAV flights without straining his neck.

The beauty of a North Slope afternoon.

Off

Traditional

White

Please Create a Profile

Anonymous — Sat, 08 Aug 2015 05:47:32 +0000

Please Create a Profile Anonymous (not verified) Fri, 08/07/2015 - 23:47

Starting yesterday, August 6th, we began data collection flights. Because the autopilot is still on the blink, they are manually flown up to around 500m (1640ft). Unfortunately, this means that the shape of the profiles is very dependent on the wind and that the ascent and descent rates are not as consistant as is possible with an autopilot. Futher, this limits our maximum altitude to significantly less than the 2km (6561ft) originally sought. Our record is 850m (2788ft), but that is less than half of what is desired, and it is very unlikely that we will profile that high again as the aircraft is little more than a dot tracing an arc against the sky. At that range, the possibility of losing control is huge. Finally, since the manual pilot needs to maintain visual contact at all times, this method precludes any flight into, or above clouds. Since the development of clouds at the shoreline is a focus of this research, there is a significant handicap in operating manually.

Fortunately, our friends back in ��«��Ƶ have not yet forgotten us, and were kind enough to send a series of packages with the hardware necessary to reliably regain manual control from the autopilot. Since the autopilot drop outs occur between 35m and 175m, the current plan is to manually fly the sections of the profile below 200m, and hand off to the autopilot above that level. This will allow the profiles to continue through clouds, and hopefully up to 2km. From repeated testing, we know that zoom climbing up to 200m in a wings level attitude and a normal descent from 200m in a wings level attitude provides repeatable success at transiting the "danger-zone". Of today's seven profiles flown using this method, all seven made it though the altitude band in question with zero lockups. We're hoping for the same level of success with the autopilot connected, but new hardware will let the pilot take over even if a lock up occurs.

We're very much looking forward to getting the autopilots up and running, but for now, capturing any data is great step forward, and we'll continue the hourly 500m profiles as weather (and the bears) allows.

Our resident polar bear confirms that the Arctic Ocean is, indeed, cold.

Preparing for a profile from our mobile groundstation.

DataHawk off the bungee!

Off

Traditional

White

Running Interference

Anonymous — Thu, 06 Aug 2015 07:52:53 +0000

Running Interference Anonymous (not verified) Thu, 08/06/2015 - 01:52

While we managed to avoid damaging any aircraft today and brought a 6th plane back to flight status, that's about all for this 3rd day at Oliktok. The first flight showed that the changes made last night solved the pitot/static isssue, but did not keep the autopilot from hanging. By rigging an aircraft with a dedicated RC receiver to control the servos and monitoring the status of the autopilot with the groundstation, we were able to maintain control of the aircraft during all of our flights while testing the autopilot in flight. Multiple test flights showed that the autopilot would run reliably at low altitude and at high altitude, but that around 50m above the ground the autopilot would often hang. It never took place at a flight time consistant across different flights, nor could the problem be replicated on the ground, even in flying mode.

Since the drop out occurs at such a regular altitude, radio interference, especially from the long-range radar on site, seems to be a likely explaination. To hopefully counter this, the entire payload bay, and most of the electronics, had a foil shield applied. A few test flights after dinner showed that the shielding did not work, and that the autopilot still froze up. Unfortunately, this put us almost back at square one.

Luckily, we have a few ideas about antenna placement and telemetry frequency which we look forward to testing tomorrow.

In conclusion, today was generally slow day, but we did manage to get some excellent flight tests and collect plenty of data about the issue.

Visitors on the flight line!

Gijs launching the DataHawk

Nathan and Gijs prepare to launch the DataHawk in the shadow of the suspicious radar dome

Off

Traditional

White

Good Weather, Good Progress

Anonymous — Wed, 05 Aug 2015 07:50:30 +0000

Good Weather, Good Progress Anonymous (not verified) Wed, 08/05/2015 - 01:50

We've gotten into the air! Thankfully, the weather cooperated today and we ended up with some gorgeous flying weather. Blue skies, calm winds, and no mosquitoes make flight operations a breeze, and were very much appreciated after the past two days of rain, wind and fog. We managed to get approximately 10 short flights across 5 planes, but unfortunately, that was all. An issue with the autopilot caused the pitot/static system to become unresponsive and freeze displaying the last recorded airspeed. Typically, this was accompanied with a descending spiral with increasing throttle as the autopilot attempted to maintain the commanded airspeed. This, by itself, is enough to force us to fly the aircraft exclusively where human pilots can see it well enough to fly it out of danger, but very often the pitot error was accompanied by a complete and total loss of control, both through the groundstation and RC override. This does not result in a very pretty arrival with the ground, and unluckily, it had to happen more than once to show that it was a problem present on all craft. At one point, only 2 out of a total of 7 aircraft remained airworthy, but because of its superior design and construction, the DataHawk is frightfully tough. Thanks to that, we'll be starting tomorrow with 5 of 7 operational, and the prognosis good for the other two.

As for the autopilot issues: we have the very best SUAS minds this side of the Brooks Range working deep into the night in hot pursuit of the problem. Besides, what fun is it without a good challenge?

There's a bear in the air! Nate piloting the DataHawk Gijs watching the DataHawk fly over

Off

Traditional

White

Greetings from Oliktok!

Anonymous — Tue, 04 Aug 2015 04:55:28 +0000

Greetings from Oliktok! Anonymous (not verified) Mon, 08/03/2015 - 22:55

After arriving at the Oliktok Point Long Range Radar Station (LRRS) around midday yesterday, we spent the afternoon at the atmospheric observation facility (called the AMF) preparing the small fleet of DataHawk aircraft for flight, and the evening getting oriented at the LRRS.

In the morning, after a quick breakfast, we piled into the pick-up and returned to the AMF to finish preparing the aircraft. Although the physical assembly was complete, a software issue in the autopilot causing loss of RC link forced us to break for lunch without completing radio range checks. Work on the autopilot took place after lunch. Running several simulations of proposed flight paths tested the paths, stored them for future use, and provided training for the crew. Despite the continuing rain and fog, the forecast showed potential for flight appropriate conditions after dinner. Upon reaching the AMF in the evening, half of the crew took the plane in the truck and drove east toward the oil rig on the point while the other half monitored the radio link status. At 1.4km, the radio control link remained strong, but the telemetry link had dropped to -85dB. Given that the aircraft was on the ground and in the truck, we anticipate range to be significantly greater than 1.4km. Unfortunately, the weather had not improved as forecast, and flying was pushed off to tomorrow.

After range checking, we took time to calibrate the IR sensors on the top and bottom of the aircraft. Using a FLIR infrared camera to provide a temperature reading, the aircraft was positioned over several different surfaces to obtain temperature data. This data was used to generate a curve that will be used to interpolate temperature. In flight, the IR sensors will be used to determine the temperature difference between land and sea.

In all, things are going well here, despite the delays in getting aircraft in the air. Given the forecast, flying tomorrow is very likely. We are looking forward to at least testing the systems before moving on to full profiles.

Off

Traditional

White

Look ma, no hands!

Anonymous — Fri, 26 Jun 2015 06:02:49 +0000

Look ma, no hands! Anonymous (not verified) Fri, 06/26/2015 - 00:02

Now that the EA-DDDAS framework has been put through the paces, we're planning two days of flight testing that allow the system operate fully autonomously. This means that the high-level planner and trajectory optimization layer will be left to their own devices to communicate, generate energy-aware trajectories, and send seamlessly stitched trajectories to the aircraft in real-time. The goal is to have no user intervention once the system has been launched and a goal has been set. We're planning to use the decommissioned Reese airfield to conduct flights upwards to 1.75 km away from the ground station using stationed observers who are always in communication with the pilot in command.

One of the recent challenges faced in the field was the ability (or lack thereof) to continuously send new trajectories to the PixHawk autopilot without causing the aircraft to RTL (return to loiter). Since the planners are generating paths ahead of the aircraft position, a method was needed to feed the trajectories to the aircraft with the correct timing. An additional constraint was the number of waypoints that are manageable on-board the aircraft. Since we use a 900 MHz radio link for long distance command and control, data rates are slow and packets are often lost during transfer. We found that a 100 waypoint plan is workable with the given data rates we see over the 900 link. With this in mind, we designed a circular buffer mission plan algorithm that allows trajectories to be continuously updated to the aircraft. Using a partial write command, we update certain ranges in the 100 waypoint plan. If a plan goes beyond the 100 waypoint limit, the plan is broken into two plans that wrap around back to the 1st waypoint using a DO_JUMP command. While this works well, it introduces another issue. If plans are being generated too quickly for the aircraft to follow, a queue will develop. To prevent this queue from over-writing points that the aircraft still needs to fly, the MISSION_CURRENT MavLink message is checked to see where the aircraft is in the circular buffer. This provides the necessary logic to delay the next waypoint write so only old waypoints are overwritten!

Sorry for the lack of generated optimal trajectory pictures so far, we haven't had a chance to dig through all of the logged data. Once we're back in CO, we'll be sure to have some results up ASAP.

Off

Traditional

White

Successful System Test!

Anonymous — Tue, 23 Jun 2015 19:29:50 +0000

Successful System Test! Anonymous (not verified) Tue, 06/23/2015 - 13:29

We've flown a full system test with data flow originating from the dual-doppler radar synthesis, feeding the energy-aware path planner resulting in plans sent accordingly to the PixHawk autopilot on-board the Tempest aircraft. Several different radar and atmospheric planner configurations were tested today to see how the aircraft performance varied with the different types of wind data. Data is currently being processed but some more of the interesting trajectories will be posted later this week. For now, check out a POV video of our Tempest aircraft (the loop was NOT an energy-optimal path!)

[video:https://www.youtube.com/watch?v=W5pUZ4PWlMk]

Off

Traditional

White