“Science” Agriculture AP in Idaho

[kd7ost]

This is the comment thread for the working agriculture thread found here.

http://www.rcgroups.com/forums/showthread.php?t=689974#post7507769

Dan

[quailbird]

Dan, I really enjoyed this article. I had prayed for you several times and, boy, did you do a good job! The Lord quite often, on certain things, will say to me, " You an't seen nothing yet"! So expect big things!

[kd7ost]

Thank you Carl,

Perhaps people like you are the biggest blessings to us all. ;)

My page three (picture 5) was posted and it was the wrong shot. I changed it out to one that was more clear and is easier to read.

Dan

[skymind]

Well done Dan! Hopefully the FAA opens this back up for you.

[Real Ira]

Dan,

That is such a productive application of RC/AP.

Your work could really make the difference crop yields that results in food getting to hungry people!

If this is left to the big players it will be priced out of most farmers reach. :(

Ira

[kd7ost]

That's what I've been saying all along. A retired guy that is an RC pilot can do a great job and supplement their income when the work is in season. I had a banner day once last year making $1,300.00 in 6 hours shooting 13 fields. I question whether a corporation can dispatch a pilot from afar and shoot just one field for that price. And be there the day the crop advisor calls you up. That’s good periodic money to a retired guy sitting at home. But it’s chump change to a CEO. If we get lobbied into obscurity, big business won’t take over the Ag AP. Instead it will prevent AG AP to all but a few cases where expensive litigation is involved. It would stop the activity and not enhance a corporations income.

Dan

[Real Ira]

Well with the good publicity you may at least get calls from "out of country" to do some shoots.
It's funny. I always though this was the home of opportunity and innovation and now I see that being squashed and the mantle going to some of our European friends.

Their governments seem to see the potential for this work springing from and growing from individual entrepreneurs and here there being shut down? :mad:

It's actually kind of embarrassing. :o

Ira

[kd7ost]

Well you and I aren't in the same camp as for as handling this. So maybe we should just stick to the topic of Ag shooting and its value. Many people won't post if we turn this into an anti FAA thread.

Dan

[skymind]

I was just looking at Colorado DoA Crop reports and was surprised to see potato's listed, 2,404million lbs harvested in 2006 (vs 3,809million lbs of spuds for Idaho). Is this activity of Ag business getting involved in Ag-AP relatively recent? I know that for years various Department of Agricultures have issued contracts to flying outfits to do AP work over large areas and higher altitudes. But closer field-by-field inspection to not only estimate individual field yield but see if and where issues need addressing I wasn't aware of.

I actually just talked to the FAA regarding use of full-scale experimental aircraft. It's a total no-go with a certain exception. That explains why I recently saw a Long Ez put up for sale by a NorthWest U.S. AP operation. I guess they learned about the rules <after> they bought it and now need to dump it.

Tom

[kd7ost]

I've been reading articles about using RC aircraft on and off for a few years in Agriculture. But what I have read is from University researchers. In my opinion, (and pardon the pun) the task is ripe for the picking. There is plenty of evidence that proves the pictures can be taken quickly and reliably. That’s not the issue holding it back from wider spread use. We can supply the high resolution pictures. But there needs to be a middle process where someone works with that picture to make sense of it. I geo reference the photo. That’s an easy thing to do with a handheld GPS. But then someone skilled with GIS needs to get involved. They need to add a layer on the picture to identify Lat/Lon and area of coverage. Working with the initial identified problem, weed or insect infestation, watering issues, plant stress in general, the crop advisor or farmer now knows exactly how much product to buy and where it needs to be placed. Treatments are expensive. But increased crop yield can cover the cost and a lot more.

The disconnect at this point is having the precision Ag manager with a GIS technician in the process. I got lucky in this market because Simplot had those tools in place and the AP was the only missing element. They came to me. In most cases it’s the other way around. I do strongly feel that the potential for growth hasn’t even been realized let alone tapped. Once companies realize the potential benefit, they can put the GIS guy in place.

Potatoes are not Idaho’s cash crop. It used to be. We are the Potato state after all. But the cash crop is Onions and that’s where I do most of my work. I shot some fields for a Dupont tech two years ago. Between fields we stopped for lunch and talked about Onions. Well he talked and I listened. He told me that in a 600 acre onion field, let to grow on its own, that the yield is small and doesn’t weigh much. The onions are sold for processing. They become dehydrated parts in packaged foods. Take that same field and spend thousands of dollars a year increasing Onion size and weight of yield. The prime product is sold to Restaurants and grocery stores. Blooming onions like what you eat at Outback steak houses and what have you are the end result. The sales difference between the first crop and the second crop can increase by as much as 1 million dollars. I asked him to repeat that several times because I thought he was pulling my leg. But he was serious. I still don’t understand crops but to me that told me there was a tremendous advantage to precision agriculture and crop management. It’s a huge potential.

Dan

[skymind]

Well this entire topic makes an effective sales & recruitment pitch. It's certainly got my interest up and I have other, legal, options for aerial platforms.

[kd7ost]

You should go for it. Satellite and Full scale have been the norm. Satellite is time consuming to get and lacks the resolution we can get by a long way. It provides about 1 meter per pixel where we can provide down to a couple of inches per pixel depending on altitude and camera used.

Full scale verticals have typically been expensive and not many pilots do real verticals. The oblique isn't that valuable because it blows the scale of things way out and you can't get accurate measurements without a lot of work. The vertical is fast and easy.

Dan

[skymind]

I can actually help put you in the position to do the same thing yourself, assuming your position with the govt doesn't preclude moonlighting. I tend to think the pricing of full-scale is based on their success finding clients who will pay those rates.

[Bill Harris]

Wonderful article. You've hit the niche similar to that which I was working on when the Federales intervened....

Should your new nickname be Spud? :D

--Bill

[kd7ost]

I can actually help put you in the position to do the same thing yourself, assuming your position with the govt doesn't preclude moonlighting. I tend to think the pricing of full-scale is based on their success finding clients who will pay those rates.

I appreciate the offer but no,

The government job has no problems with it but I am gone an awful lot. I put in many hours and don't need the money from AP. At this point I want to see what I've worked on so hard for the last 4 years be something that we as Americans can do. I'm not worried about taking pictures right now. I'm worried about taking pictures in the future when I move to a job that is less time consuming. Or when I hit retirement and want to go fly for a few bucks.

Dan

[kd7ost]

Wonderful article. You've hit the niche similar to that which I was working on when the Federales intervened....

Should your new nickname be Spud? :D

--Bill

Thanks Bill,

Maybe "Tater Tot". :D

Or for now, "Tater Not". :rolleyes:

Dan

[Real Ira]

Well you and I aren't in the same camp as for as handling this. So maybe we should just stick to the topic of Ag shooting and its value. Many people won't post if we turn this into an anti FAA thread.

Dan

Your right Dan.

This subject sort of touches everything we do any more but this forum is really not the place.

Ira

[skymind]

Dan, Here's a few questions for you.

1>Is it practical to do this work largely on a wekend only basis? Getting calls and scheduling the sortie for the weekend?

2>How much do cloudier days impact the work?

3>Is the client always the owner of the field?

4>Do you vary your altitude based on field size so that you try to get the entire field to fill the frame?

5>Do you ever stich the photos?

6>Have you ever had to bring your vehicle down or low or reposition due to manned aircraft in the area?

Tom, off to the salt mines

[kd7ost]

Dan, Here's a few questions for you.

1>Is it practical to do this work largely on a wekend only basis? Getting calls and scheduling the sortie for the weekend?

Sure it is. I've done a lot of my work on weekends only. But at times it wasn't enough.

2>How much do cloudier days impact the work?

Depents on how cloudy. I set my camera for manual mode and shoot at 1/1000 and 1/800 of a second. On bright sunny days the pictures are light. On cloudy days they can come out dark. I just lighten them in a photo program.


3>Is the client always the owner of the field?

Not in my case. Simplot has a lot of local small district offices. I've never shot a field yet where I wasn't hired by Simplot. The land owners in many cases will hire a company to supply a crop advisor. These guys are skilled agronomists. It's a real science. They know the chemicals, the results of plant and soil testing and how to fix problem area's, etc. The land owners pay the crop advisor to help them increase yield. The payoff is when it's harvest time. When the Land owner sees it pays dividends to hire a crop advisor, it's an easy thing for me. I don't have to solicit to the Land Owner. I just take the calls and go shoot.

4>Do you vary your altitude based on field size so that you try to get the entire field to fill the frame?

Yes.

5>Do you ever stich the photos?

I haven't had to yet but it might be a part of the future process depending on altitude limitations if any are imposed.

6>Have you ever had to bring your vehicle down or low or reposition due to manned aircraft in the area?

After flying my plane in excess of 2,200 miles, (Logged by the on board GPS) there are only two instances I recall where I've had to do that. The environment as you realize is pretty wide open. Farm land is not mountainous or full of wooded area's. You can pretty well see the horizon in all directions. Since the weather has to be pretty cooperative, you're flying on mostly clear and fairly calm days. You can hear GA from many miles away. In most cases you can hear them before you can even see them. As much as 10 miles or more. The two times I recall had a plane coming towards the operation from in excess of 5 miles. In both cases I landed the plane and had to wait for the GA to pass and move off. We get a lot closer to GA at the AMA flying field I can assure you. Not that we get close there mind you.

Dan

[skymind]

Thanks for all that Dan.

[HELModels]

I would like to do this too. I've taken some necessary steps.

1) learn to design, build, fly
2) learn to make molds
3) learn about GPS and simple autopilots

4) not there yet - need to combine and expand above steps and enlarge my design

There is alot to it and you have really promoted this excellent use of AP.
Agriculture usually isnt in densely populated areas which makes AGAP all the more freindly.

Harry

[kd7ost]

I would like to do this too. I've taken some necessary steps.

1) learn to design, build, fly
2) learn to make molds
3) learn about GPS and simple autopilots

4) not there yet - need to combine and expand above steps and enlarge my design

There is alot to it and you have really promoted this excellent use of AP.
Agriculture usually isnt in densely populated areas which makes AGAP all the more freindly.

Harry


Thanks Harry,

I'm completely open about everything I do with AG AP and how I do it. I want this to succeed everywhere. Please feel free to ask me any questions. I'll do everything I can to answer what I can.

Dan

[HELModels]

Thanks, Dan.

How did you tap into the camera and then feed the shot through the video?

Harry

[kd7ost]

This plane is more elaborate than a second one would be. First off the Plane has two cameras and a video switcher. That nose turret shown in the article houses small board camera with a lens that closely replicates the field of view as the digital Camera. I'm using the Fuji E510 for the stills since it has a video out port and an effective 28mm lens. Wide angle prevents you from having to fly higher. The Fuji E510 Camera is no longer made but there are others that will do the job.

I position the nose camera with a servo so I can rotate it as I get over the field. I point it forward to see the field and get lined up. The same pulse I'm using to move the nose camera activates the video switch. I have it adjusted to activate the Video switch and put the digital camera on line just as I'm nearing the position. Here's a poor quality video showing those two camera's in operation. I don't have it adjusted perfectly in this video but it still gets the job done. I also carried too much power and didn't get a real good vertical shot. The two camera's go into the video switch and the output goes to a 600 mw 2.4 GHz AV transmitter sending the signal down to me so I can see what the camera sees.

http://media.putfile.com/Typical-Ag-Mission

Future designs will have a single digital camera in the nose. I'll be able to rid the second camera and video switch.

Here is another thread that discuss the process.

Dan

http://www.rcgroups.com/forums/showthread.php?t=571185

[Gray]

Excellent article Dan. Well done.

Gray

[kd7ost]

;) Thanks Gray.

Dan

[skymind]

Dan, I wanted to batch up my questions and I've got a batch of 3.

Would it be true that not all crops have the same level of opportunity? Corn, wheat, orchards would seem to have lower levels... Onions are greater. Anything else?

Have you gotten feedback from readers of the magazine?

Does time-of-day matter much for the client of the photos?


Thanks
Tom

[kd7ost]

Dan, I wanted to batch up my questions and I've got a batch of 3.

Would it be true that not all crops have the same level of opportunity? Corn, wheat, orchards would seem to have lower levels... Onions are greater. Anything else?

I take it by level of opportunity you mean value of photo versus value of crop? I think much of it depends on what an area’s local cash crops are. I live in the potato state but more potatoes are grown outside of Idaho for example. Everywhere from the Netherlands to China and Russia can produce potatoes like Idaho does. So Idaho doesn’t maintain a market lead in Potatoes. But Onions from here are not yet replicated elsewhere that share the value of Idaho Onions. Development is moving along though and it’s just a matter of time. Corn generally does have a lesser value as a cash crop. Sweet corn for human consumption is higher value than feeder corn for cattle. Processed corn, for canned and processed food is somewhere in between. But feeder corn is the most prevelent. The value in all the corn field pictures I’ve taken comes from some sort of human induced issue. Like in the magazine pictures. One picture on the last page had to do with a sub contractor having a broken nozzle on his fertilizing rig. The land owner was considering holding Simplot liable because they are contracted as a crop advisor for the land owner. In that case the picture was paid for by Simplot to confirm suspicions that the company they hired to fertilize had defective equipment. It allowed Simplot to take the picture to the Sub contractor and prove that they had an issue and where to place blame and how to manage a fix action. Sometimes the pictures are used in a different way than just looking for weeds. I would imagine grape vineyards would be of high interest in places where wine is a staple of the local economy. I’m not sure if I understood this question so if I didn’t answer it let me know.

Have you gotten feedback from readers of the magazine?

I have at that. It surprised me. I left out any way to contact me in the article. I knew with the FAA debacle coupled with my day job, and the start of the US fire season that I wouldn’t be able to do much of anything. Between family and job, my summer is pretty filled up. I would rather be scarce than unreliable.

This is good though. I figure the people that take the time to track me down must be serious. Due to my busy schedule I’m only doing consulting. People want planes and I don’t have the time to build. One guy was approached by the Potato Growers of Maine and they would like to see if he can do for them what I’ve done for Simplot. He should have a handle on it real soon. He owns this. http://www.scalercmodels.com/
He’s a high level modeler. I’m losing opportunity but I have to. Also at this point I really want the process to grow. Big business hiring modelers to do AP have tons more weight in influence with the FAA than I do as an individual. And guys that read my posts know I’ll do what I can to guide anyone to success at this point. With millions and millions of acres of crop land in the US, the market isn’t even tapped let alone competitive.

Does time-of-day matter much for the client of the photos?

Thanks
Tom[/QUOTE]

I don’t know. I’m sure there must be a best time and a worst time. I would imagine it depends on the type of crop, the type of imaging, and what you’re looking for. By its very nature, most all my shots have occurred between 10 am and 4 pm. It’s usually the best time to connect up with the right crop advisor after traveling to his district. It doesn’t take long to get the shot. Too early or too late would have the sun low on the horizon and cast shadows as well as not being as light on the crops themselves.

Dan

[HELModels]

I would think that crop growers and managers, scientists would know exactly at what stages they want imagery. I would also think that any grower who requires application of some product, even if it is just water, would want to increase accuracy and efficiency, or know if their attempts to do so are working.

I'm still working up to a large enough platform to handle the role.

Thanks for all the info.

Harry

[skymind]

I would think that crop growers and managers, scientists would know exactly at what stages they want imagery. I would also think that any grower who requires application of some product, even if it is just water, would want to increase accuracy and efficiency, or know if their attempts to do so are working.

I'm still working up to a large enough platform to handle the role.

Thanks for all the info.

HarryAre you thinking as large as Dan's?

[HELModels]

That is a good question, I was thinking airframe large enough to handle the gear. Dan has hinted that technology has shrunk enough that his next workhorse wont need to be as large. I have a 6.5 foot wing, a pod design which can be upsized, a basic overall design. It is no small or inexpensive task and the equipment will dictate.

I defer expertise on any of this to Dan, of course, and appreciate all his pioneering. Maybe he has some firm sense of how large an airframe is really necessary and might share that, if he hasnt already. I dont think a slowstik is gonna cut it, but I could be mistaken.

Harry

[kd7ost]

This could be a lengthy response if I take the time to cover the bases. It's wide open I think but I'll list what drove my choice and outline the pro's and con's as I know them to be. I'll write up something in a word doc and post it here.

Dan

[kd7ost]

The size of my plane was driven by several factors.

First was size of optical sensor unit. Simplot asked me if I could make a plane that would carry a camera that weighed up to several pounds so they would have flexibility with multi spectral cameras.

Next I had to decide on gas, glow or electric. Electric wasn’t an option at the time due to sheer cost of a motor, speed controller, batteries, charger, balancer etc.

I left glow out for a few reasons. Fuel residue was one although with a pusher design it’s not a big issue. Glow fuel is expensive. Glow fuel must come mail order or from the LHS. Glow engines use glow plugs and they wear out fast when compared to a spark plug. On the plus side, glow engines have superior power to weight ratio over gas engines. Glow engines generally require a starter and battery though so more field equipment is required. Oh, and glow engine having smaller pistons for the power to weight ratio coupled with higher rpm create vibration that is easy to deal with compared to gas. High frequency low amplitude vibration is easier to work with than low frequency high amplitude.

I knew I would have vibration issues with the gas plane but it gave me some options I preferred. It uses a spark plug and they don’t cause dead stick landings like a glow engine might. I could buy gas in any little town USA and mix it with common 2 cycle oil to keep flying. I could get a gas engine with a magneto ignition so no batteries were needed to create spark. I bought the smallest engine I could find that featured an optional spring starter. I bought the Fuji BT32A. At almost 2 cid it was a workhorse and would provide plenty of power. Now I could run in the field all day, start the engine by hand, and it gets great economy too.

The possible, yet unknown future camera options and the engine size were the two driving factors for size of plane.

But the low frequency high amplitude vibration caused me fits getting the Fuji E510 wide angle lens camera taking clear pictures. The engine was soft mounted with stiff rubber mounts, and the camera mount itself took a bit of finessing to get tamed down. That was the hardest part of getting the plane flying. I spent several months building and testing new mounts before I finally hit on the working combination. It was a simple fix too.

With the gas engine, I get 1 hour of full throttle with 1 quart of 2 cycle gasoline mix. With the magneto and spring starter, all I have to do is fuel up and go. The plane uses two battery packs for receiver and servo. I built my own packs with 2400 mah NiMh batteries. These would give me about 10 hours of flying without having to recharge. NiMh also don’t have a memory effect like NiCds and can be fast charged. They do self discharge faster in storage so must be charged before heading out.

Well, after all was said and done, the most fields I shot in one day was 13. I was always ending up being conservative in my design parameters as I had more battery power than I needed. (I like that actually) The larger 3 to 5 pound camera’s never emerged. My Fuji camera has done all the work.

I’ve been following Dave’s thread on his 22 pound electric conversion here. http://www.rcgroups.com/forums/showthread.php?t=692179

That having been said I strongly feel smaller electric planes would be great. The cameras are small but I would definitely go with a 28mm lens. You need altitude to get the field and cameras with narrow FOV’s mean you have to fly higher. I fly to 3,000 feet to get a 600 acre field with a center pivot irrigation system. Most are not that high though. You have to have room for the video system with its battery, and I like to use a GPS return to launch unit. I also use the co-pilot to stabilize roll. I fly with rudder input once I’m up to keep the camera from swing all over if the plane were to roll as well as yaw. I put the engine or motor with a few degrees of up thrust. This way the plane will climb on its own under power. Once you get up high its too easy to over do the elevator and drop airspeed and stall causing a loss in altitude. I set the plane up so it will do things for me that I don’t have to do. When the camera is aiming down, you lose visual in the downlink with the horizon and you can’t tell if you’re going up or coming down. Being able to just fly up by keeping the throttle full is a big help.

I think I can do most of the work I did last year with my 6.5 pound Dragonfly design. I have the wrong camera in the nose but if I put in the Fuji or another with 28 mm it would do fine.

Naturally a bigger plane is easier to see at a distance and handles windy conditions better. I’ll post a picture of the Dragonfly and include a link to the rather extensive build thread on it. I like to move the camera in flight from horizontal, (front looking) to see where the field is and get lined up. Then be able to rotate it to vertical for the picture.

http://www.rcgroups.com/forums/showthread.php?t=573747

I think there are many ways and platforms to pull this off.

Dan

[skymind]

I wondered about the specific altitude you used, 3000ft is way up there.

In your Pegasus thread, do you described how you integrated your GPS to the camera. If so you don't need to re-post it, I'll go thru that thread anyway.

[kd7ost]

I wondered about the specific altitude you used, 3000ft is way up there.

In your Pegasus thread, do you described how you integrated your GPS to the camera. If so you don't need to re-post it, I'll go thru that thread anyway.

I would be happy to reply to that here. There have been so many posts in weird locations on the AP thread that I don’t think I could find anything specific there about the Pegasus1. This forum allows things to remain consolidated so if you guys have any questions please feel free to ask here and I’ll build this up. Any other solutions from other AG pilots can go here too. Like they say, there’s more than one way to skin a cat. Even though I have complete success with AP from my AG plane, I wouldn’t replicate the original design without some changes. I’ve thought of better ways to do things.

In the Video Overlay on the Pegasus1 I use the ICcircuits OSD-GPS or the OSD-GPS (ID). Either will do fine but there are other units out there that will work too. I don’t find that all the information on that display is useable to me in my operations but the extra data isn’t a problem either. Its plug and play after you put on some connectors so integration was a breeze.

http://www.icircuits.com/prod_osdgps.html

It’s a 12vdc unit as is my microphone. I use a small 3 cell Lipo (1200 to 1500 mah) to power the whole video system. Some of the video system, like the 600 mw AV transmitter are 5 volt units. So I take that Lipo voltage through a power switch. The voltage goes right to any 12 volt device and also to a 5vdc regulator from Dimension Engineering. That regulator is a standard form factor with .1 pin spacing. It plugs right into a servo lead for ease of use.

http://www.dimensionengineering.com/DE-SW050.htm

I also find this to be helpful. I buy color coded RC servo extensions from here.

http://electrodynam.com/rc/conn/index.shtml#ext

There are a few reasons for this. One is that they are just so robust. They equal top of the line JR gold plated pins and connectors. They have large wires to prevent current drop in heavy loads and the shells fit tight. I’m sure we’ve all seen some combinations of cheap connectors that slide in and out so easy that you can’t help but wonder if you want to trust them for long term. The gold plated pins stand up to the test of time. And, they can be ordered with various colors. I take advantage of that color coding. I use Yellow ones for all my GPS signals Red for 12 vdc power, Green for 5 vdc power, and Black for my standard servo connections. This makes it easy to trace signals through a busy wiring harness. There’s no right way to do this. It saves having to put labels on things and keeps the harness neat and clean.

The OSD unit comes with connectors but has a good diagram showing how to connect power, GPS and video in and out. I use a red RC connector to put 12 volt power to it. Just the red and black wire is used. I use a yellow RC connector to bring in the GPS RS232 data. Then I use a molded end RCA wire to connect video in and out. Other wiring with smaller connectors will work on the video too. I just have so much room in the Peagasus1 that it’s easy to use the RCA jacks.

Video “out” from the camera goes into the video “in” on the OSD. Video “out” from the OSD goes to the AV transmitter. That’s where I add in the RCA jacks for standard video connections. The GPS signal comes from my GPS and goes to a Y connector. You can easily drive several devices with the GPS output. The RS232 signal has a good “fan out”. This means it has enough power to drive a handful of devises at the same time. One of those split outputs goes to the GPS in on the overlay board.

Once the GPS is on and has lock, the video systems are on and the OSD has power, you should see your normal video link on your monitor, but the OSD has decoded the GPS information and laid it on the video signal.

As a side note, the other GPS output goes to my “Return to Pilot” system. I use an obsolete UNAV PDC-10 controlling rudder. The OSD and the UNAV products use the standard NMEA 0183 at 4800 baud signal from the GPS. This signal is an industry standard.

I like to keep a handle on everything in the build/integration process by working in one system at a time. I consider the flight control and wiring, power, receiver, servos etc to be the first and most critical part of the integration. I do all of that first and then fly the plane and work on setting up, control throws, etc till the plane handles like I want it to.

Then I add in the GPS return to pilot system. It’s a layer that goes into one part of the original RC system. Then I work on that to get it operating like I want.

Finally I integrate the Video layer into the aircraft package. For the most part it’s a stand alone and separate system. Approaching it in a systematic way like this helps improve the ability to manage the separate systems. It keeps the pieces easy to keep a handle on. You need to be the builder, designer, integrator, pilot, troubleshooter etc. Doing one system at a time before moving on prevents you from having to wear too many hats at once. Things can get pretty daunting if you have problems in several areas. Give yourself the best chance of success and always make sure the plane flies in your comfort zone first. If all else goes wrong, you can land without having a Tiger by the tail and it will be in one piece so you can resolve what you need to and try again. Always put the plane handling ahead of everything else. And as you start working the GPS or video system in, and you detect you want to change something in roll control, pitch, thrust whatever, drop the GPS and video stuff to tweak the handling. This will keep you in control as you have to change all those hats. The plane will be well dialed in and you will know each part very thoroughly before moving on.

Dan

[wattnoise]

The history channel mentioned Simplot last night on their series "The States" in the sequence about Idaho... Kind of an odd series if anyone has been watching it - they seem to expound upon certain things that are not necessarily key to the history of the particular state... An example was quite a long sequence about Teddy Roosevelt and the Rough Riders in the North Dakota series... ?!?! A majority of RR's were recruited from AZ, NM, and OK!... Sorry for the OT - but, it just always makes me wonder about the validity of other info from resources such as the History Channel when a known "inaccuracy" is found...

Anyways - back to a crop story, too... Corn may be on the rise... The story out of Mexico is that many agave farmers (tequila) are switching to corn because of the expected upswing in demand for ethanol... Margaritas will be more expensive in the near future!... ;)

Kevin

[kd7ost]

Hey, I caught that show by accident. Was channel surfing in the shop and scanned past it just as they were showing the Simplot Potato process.

Dan

[kd7ost]

To add information to the video/OSD reply above,

I always downlink Audio too. This is easy as all the AV systems I use have an audio channel built right into it. You can take the audio signal from your camera’s output and put it right into the AV transmitter.

Ever notice when you’re flying regular RC, when you get high or far away you can go from full throttle to minimum throttle and it can take a second or several before you hear it transition on the ground. The speed of sound in these cases shows up. By down linking the audio you can hear transitions and camera shutter sounds at the speed of light. You get what is perceived as instantaneous feedback.

One way to tell if your plane suddenly starts dropping altitude is an increase in pitch from the throttle as the prop unloads.

I’m going to repost a link to an old video. Bear in mind I don’t care about video recording so this isn’t the best quality. But it gives enough of a visual to show you what I have going on. In this video here is what is happening on the ground. What I’m doing to fly this test and demonstration flight.

I start the take off run. Once up to speed I pull up elevator to rotate. I keep a little bit of up as I’m watching the plane to ensure I clear the power lines along the rural road. Roll is being managed by the co-pilot which I don’t put on a channel to enable or disable. This way it is always enabled. I don’t need to put in aileron control unless I want to roll. Then you have to stand on it. Anyway, the engine thrust is a few degrees up. It’s enough up thrust that under full power the plane always climbs. This was part of that aircraft setup I talked about in the last post. As the plane clears that rural road, I’m not even using the right stick anymore. The plane has been set up by me in its response characteristics to make my job easy at this point.

Soon you see the nose camera begin to tilt and the plane begin to perform a flat turn to the right. I already knew what line of fields I needed to center up on to shoot the field I wanted. At this point I have transitioned to flying by watching the monitor so I could see what the camera sees, and start to position it center of that line of fields.

Note: There is an audio link so I can hear the engine. You wil also hear some spaced beeping that tells me my Basic Stamp micro controller is bumping the camera shutter just far enough to hit the first shutter detent on the Fuji E510 digital camera. This is my “watchdog” and the beep tells me my camera is being prevented from going into sleep mode while I’m not looking through it. I know when I switch over to it, it will be awake and ready to shoot my field. Also, the video overlay shows me altitude as well as other things. All I’m looking for here is to see my altitude is increasing. I don’t care how high. In fact the GPS is reading Meters above sea level so it’s not too critical to me how high I am. I just want to see at a glance that the numbers are getting bigger as I look downward through the camera and can’t see the horizon for reference. With the engine sound droning constant on one hand, the GPS altitude information from the OSD gives a second piece of feedback to tell me I’m climbing and not leveling off or descending. I now have an audio and visual indication that all is going well.

At this point I still haven’t touched the right stick after take off. I’m flying with the left stick only. Throttle is full and steering is with rudder. My right hand is on the knob that tilts my nose camera. The pitch, speed and roll are taking care of themselves. The worst thing I can think of would be trying to manage those issues while positioning a camera. I can’t emphasize enough how important the plane setup is in order to ease your task as the photographer.

Next you see some switching between cameras. This isn’t very clean at the switch point. When I built this plane I had to design and build a video switch. I automated the switching process though. I use the servo signal to the little board camera turret to accomplish the switching. I built an electronic switch that was adjustable. Its switching point isn’t dialed in perfectly in this video demo though. I had just replaced the turret servo and needed to adjust the switch point. Anyway, the reason the toggling between cameras occurs is due to the Hysteresis in that switch. I’ve always just lived with it. Once I see the camera view rapidly toggling between the two cameras I know I’m close enough that I just rotate the knob for the turret camera a little farther and the toggling stops. I now see the view from the Fuji camera.

You can see some yaw happening just as I line up. There was a pretty stiff cross wind component from the north blowing me to the right of the field. But once I had the field framed up, I rotated the knob that activated the shutter servo and took the picture. You will also note at this time I had throttled back to stop climbing but didn’t throttle back enough. I was still a little pitch high when I took this shot. You can tell because the rectangular field was a trapezoid shape. In an actual AG AP mission I would have throttled back more or run another racetrack and come around again for a flatter shot. Being a demo video I just called it good and came down. I saw in the 2 second review that I had the whole field in the frame.

Next I rotate the nose camera forward and that switches back to its view. Next the video fades. I use a directional patch antenna on the ground and had flown to the edge of its sensitivity and had a drop out. There is a moment where nothing from the ground happens to change the planes flight trajectory. I reached up to my patch antenna and repositioned it to get the AV link back by aiming it at the planes location.

Next, I rotate the spare channel knob enabling the GPS return to launch system. About this time I also chop the throttle to an idle. The plane begins to descend with the throttle off and the rudders steer the plane back to me while the co-pilot keeps the wings level. (I use only roll control in the co-pilot. Pitch isn’t connected to my receiver or elevator) I still haven’t touched the right stick after clearing those power lines. The plane is balanced just a little nose down so it will find a point in descent where it comes down on its own without needing to use throttle or elevator. The plane gets overhead and I use the camera to look around and keep an eye on my surroundings.

The plane is allowed to loiter down on its own. Till its close enough to me where I want to take over. I leave the monitor and pick up the plane visually. I take over as if I’m flying a standard RC plane. With camera pointed forward in the video, you finally see the horizon shift as I induce roll, manually over riding the co-pilot. This is the first time I touch the right stick since right after take off. I disable the return to pilot system just before I take the right stick. You need to have rudder control back.

Finally a normal RC plane landing occurs. Here’s that video. Less than 5 minutes from takeoff to 1,700 feet AGL and touch down with no radical flying needed to pull it off.

http://media.putfile.com/Typical-Ag-Mission

The systems and set up are dialed in to enhance the planes ability to do much of the flight work for you. It’s a little different to learn but any accomplished RC pilot can put it together through careful set up and practice.

I also want to state that this is the way I do it right now. The components are off the shelf and can be worked into an airframes dynamic flight envelope without a lot of fuss. You just have to take a step by step approach.

The Pegasus2 is in the design phase. I plan to replicate new systems and experiences learned from the Dragonfly. I do not want to put two cameras, a video switching system etc into the next design. That will also eliminate the need for a micro-controller watch dogging the camera. Some cameras can have the auto shut off feature disabled. Not the Fuji. If you’re looking through the same camera that you’re going to shoot with, and it shuts off the video, you simply have to bump the shutter to wake it back up and start the time out sequence over. I consider the auto shut off a little bit of a nuisance but that function is secondary in my experience to wide FOV and good pictures. With the plane set up to fly itself well, it’s no trouble to bump the shutter for a second once in a while to wake up the output.

Another thing to consider is the camera physical location in the fuselage. In a modified tractor type plane you’re pretty well committed to put it under the wing at the center of lift for balance considerations. My prototype technology demonstrator was assembled inside of the Sig Rascal 110 with a Zenoah G26 for power. It was a great flying plane but was a tail dragger and it was always a chore to get to the digital camera in the belly. There was too much gear inside above the camera mount so I had to flip the plane on its top, or have someone hold the tail section up and set the plane on the spinner. I would have to sit on the ground and pull the whole camera mount from the plane to get to the camera. The plane had a narrow landing gear and as a tail dragger it wasn’t highly great with cross wind landings. I drug the wing tips more than a few times and struggled to keep it in the road after a landing with high cross winds. If I landed fast, it would set the tail down, and pitch the wings positive and bounce in the air. Those issues helped drive the design of the Pegasus1.

The Pegasus1 isn’t just a pusher to put the engine behind all the video views. It also has a small side area behind the engine in front of the tail surfaces. This makes it a lot less susceptible to getting out of shape in Yaw on those cross wind landings. It also allowed me to put that whole front end cover as a clam shell design. This made access to the gear real simple and convenient. I made my camera mount so that I could get to the memory card without even having to pull the camera out. I can also access the camera batteries for swap out because the memory card and batteries are in the same compartment under the same hatch.

I had to work around the nose gear too. It’s not way out front but is set back a little ways. If you put a camera in the nose you need to ensure that when oriented for vertical pictures you aren’t getting nose wheel into the frame if you have a nose wheel.

I had a custom main landing gear bent up for me that is very wide. The landing gear is set so that there is about 1 degree of negative incidence in the main wing when on the ground. When this plane sets down, it sticks and stays there unless you rotate it. You don’t want to over do that angle or you’ll have trouble rotating on take off. The Fults Tooling nose gear allows me to fine tune that angle. I use expo on the rudder/nose steering so a take off run is smooth with steering corrections not being radical around center, but it still allows me full travel with full stick deflection to turn around on a road.

Anyway, a lot to chew on here for now. I'm presenting this both as a record of what I've done, learned and changed, as well as help to anyone getting ready to go through this process. I want to state over and over that this is only one way and there must be many variations that will still net success. Feel free to experiment, come up with different ideas and post them here.

Dan

[skymind]

Wow Dan, that is alot to chew on. I have watched that video earlier and wondered about the beeps. I was surprised how quickly the flight went but as you said it was a demo.

Now I'm puzzled as to the role the GPS plays with the imaging. I thought the gps data was imbedded into the still image files in order to make the info available for the GIS work. If it's placed only onto the video (in addition to feeding the autopilot) I'm not sure what the value of it is of seeing it. Not that it's a bad thing to have it on the video. It just means, that's one thing I wouldn't have to deal with.

As for the GPS & autopiloting, you must have taken coordinates of the center of the fields so the plane will be taken there. Maybe you said you steer the plane on the 'bomb run' to the shot point, I'll have to read again and look for that. Yup, there it is. You hand steer it on the final run.

[kd7ost]

At this point I consider the OSD unit to be optional. I could live without it. Like I said above, I only use the altitude and sometimes the heading. These can be worked using other means. Like the sound. I know for example that my plane will climb with power on and full throttle. Early on I liked to see the altitude because it have me assurance. Now I trust the plane and process enough to know its climbing.

The heading is only partially useable. You would have to keep an eye on it during the whole flight, and know where you’ve been going in order to make sense of it. I fly to nearby fields so it’s not a big issue. I could do without it.

I don’t fly the plane with waypoints loaded and send the plane on its own. I tried that early on but there are problems with the GPS centering up and framing the field. It won’t do that. The GPS is reactive but the pilot is highly anticipatory. I can manage what to do in a cross wind using the downlink view much better than any GPS can do. The GPS has a range to waypoint program and if blown a little off course, it will sequence to the next waypoint as you fly past the last one. I only use the GPS as a return to pilot system and I won’t go without that unit. Interference or getting lost in an unfamiliar place and flying the wrong direction are very real issues to face. Also, having the ability to allow the GPS return to pilot system loiter the plane overhead while descending is a nice feature.

As far as geo-referencing I do that on the ground. I use a handheld GPS and get the coordinates of one corner of the field, or a corner of an intersection if it’s a round center pivot field. I also note the cardinal direction of that corner. North East, South west or what have you. I supply that coordinate with the picture so the GIS layer can be aligned with the field. Simplot does there own GIS overlay work.

I think the OSD can be left out.

Dan

[skymind]

That's all really great information Dan. Now I have a better understanding of the difference between the essential deliverable and what you use to get there. It seems a bit more doable for myself. Not necessarily having to assemble an integrated multi-component electronic system with extra circuit board and maybe custom programming too is a relief. It's alot more fun when airplanes are used for a purpose than simply ripping the sky.

[kd7ost]

Here is another document I just finished. Questions weren't asked but this is important to lay out for the record. This deals with camera shutter speeds, using screen capture from video and vibration issues. I don't know if it will be too helpful to you guys ramping up for this right now but is presented as food for thought for those future readers looking to move along in the AG RCAP operations.

Dan



Notes on screen capturing video feeds for single snap shots and Vibration

In using a screen capture utility for getting stills from an unmanned aircraft you have different issues to consider. Image blur caused by vibration and motion become more evident. Here is the reason why.

Video feeds can look very clear to the human eye as a receptor due to something called persistence of vision or POV. Even if you’re viewing a camera feed from a high vibration environment. The human eye and brain are slow to respond and doesn’t detect image changes or flashes at faster frame rates. A common example in RC is being able to use an RC optical tachometer to see the frequency of a standard light bulb. The bulb is flashing on and off at a 60 Hz rate due to the AC current polarity changes. But the human eye cannot detect that on off switching. It responds too slowly and the POV provides the illusion of the light being on all the time.

In standard US television the pictures are changing on the screen at a rate of 24 times per second. But when you view the show, your eye cannot detect those changes and the change in frames appear to be very linear and fluid. Read more about POV here.

http://en.wikipedia.org/wiki/Persistence_of_vision

I shoot pictures from my aircraft, using a digital fixed camera at 1/1000 of a second. This is a fairly fast frame rate. This serves to keep the shutter open for a very short period of time resulting in a clear image. Even then, I had to isolate my gasoline engine with soft mounts as well as build a soft mount for my camera.

Early experiments showed me that when I recorded the very clear looking downlink image, then captured one frame to view, that I had extensive blur in almost all the frames. This was because even though my POV told me the video was very clear, an individual frame was captured at a standard rate or 30 frames per second. Each individual image was basically captured at the slow rate of 1/30 of a second. Compared to my 1/1000 high speed capture in my digital still camera, the 1/30 of a second frame grab was very slow. The shutter was open long enough to allow the image to be blurred.

I have seen very clear single frame grabs from RC aircraft video systems, but only from electric aircraft. I don’t know if those frame grabs came while the plane was under power or when the motor was shut down and the plane was in a glide mode. Electric power plants certainly provide options that ICE power plants do not.

A note on vibration.

There are two factors that cause image blur all other things being equal; vibration and motion. Vibration comes when the power plant in operation induces a frequency of motion throughout the airframe. Vibration can be minimized to the airframe by careful engine selection and the use of vibration isolation mounts. Some vibrations are easier to isolate than others. Vibrations of low frequency and high amplitude are harder to dampen than those of high frequency low amplitude. Consider sounds waves as an example. Someone a few cars away from you at a stop light has a powerful stereo. You can tell because the thump, thump, thump of the bass reaches you loud and clear. You don’t hear the high frequencies as well or perhaps not at all. The material in the vehicle construction and space between the cars easily dampens the high frequencies. But the low frequency bass travels through quite well.

From the perspective of the camera itself, vibration can be further reduced by building a soft mount for the camera and carefully increasing inertia (mass) of the camera and mount till it resists motion. A soft camera mount with a very light camera isn’t as effective as a soft vibration mount and a heavier camera.

How this relates to RC planes is quite similar even if the frequencies are higher. A large single cylinder gasoline or 4 stroke engine operates at low speeds with a lot of unbalanced mass in motion from the weight of the cylinder. As you decrease the mass of the cylinder you decrease the amplitude of the vibration. As you increase the RPM, you increase the frequency of the vibration. That is easier to work with. A 2 cycle glow engine develops high levels of brake horse power for its size through high speed operation. The engine can have much smaller mass by developing that BHP with smaller pistons. Less mass means less amplitude. Since the engines also operate at higher a RPM, the frequency of the vibration is higher.

None of this means that the camera will take a clear picture even if a fast frame rate is used. It sill depends on how much vibration makes it to the camera CCD array and lens. But, it does mean that it’s easier to isolate the camera from high RPM than low RPM.

In my Pegasus design, it took a lot of work but I managed to isolate the camera from high amplitude low frequency vibration from my large single cylinder gasoline engine by use of soft mounts on the engine, an isolating camera mount, and adding lead to the camera mount to increase its inertia or mass. (Weight)

This whole document is a little disjointed but the point is if you can get rid of the initial engine induced vibration to begin with, by using an electric brushless motor for example, the chain of isolation from the power plant to the camera becomes much easier to deal with, or perhaps none at all is needed.

Also, frame rates need to be fast. Some vibration may still make it to the camera and slow frame rates will show that. Faster frame rates will capture the image before motion can make it into the camera.

Naturally there are all manner of power plants available and in my opinion they can all be worked with. Cost, turnaround time on the ground, ease of getting fuel or charging batteries must all be balanced out by the operator to select the best method for your concept of operations. If you use gas or 4 stroke engines, you will need to do the most work on vibration isolation. 2 cycle glow is less of an issue but you will likely still have to isolate to some extent and still keep frame rates high. (It’s best to use a camera with manual control settings for this) Electric simplifies all that but there is the start up cost as well as in field recharging considerations. To keep flying all day you will need to have a stack of batteries ready to go and several chargers rotating the battery packs through. Big packs will kill your car battery if you don’t keep the engine running from time to time for the alternator to keep your battery charged. There are pro’s and con’s to each method. If you only have to shoot a few fields a day, electric would be my choice for sheer ease of use and building requirments.

Using a screen capture utility to view frames on a PC will compound the vibration issue due to slow frame rates.

Dan

[skymind]

Hey Dan,

Is there a preferred or required file format they want the images in? jpeg ok? raw preferred or required?

Thanks
Tom

[kd7ost]

I just give them jpeg's.

Dan

[skymind]

Thanks Dan

[N8fromVA]

What an outstanding representation of the hobby/business. As usual, great job!

[skymind]

One aspect of this that's interesting and encouraging for this Dan is that with all the ag-imaging technologies (such as multispectral and hyperspectral systems) and services out there, that you can find a workable niche with a basic p.o.s camera that sees only in visible light or perhaps a little infrared too.

[kd7ost]

The sky is the limit. Small camera's don't always show very much. That's why I use big planes. They have a much better payload capability with ease of installation of payloads. Pilots stuck on small electrics will limit themselves to small light visible spectrum still cameras. (And pretty calm winds) Still a good thing if you're just looking at getting some hobby money. But if you want to make more, you have to think bigger than that.

Dan

[skymind]

The sky is the limit. Small camera's don't always show very much. That's why I use big planes. They have a much better payload capability with ease of installation of payloads....DanBut I thought you have been using a small point-n-shoot in your planes. I haven't compared the numbers of these camera's but I'm under the impression that a couple Fujifilm camera model (such as Fujifilm IS-1) see further into the IR band than other digitals, even if you removed their internal filter. And I'm still looking around to see what's in the market for UV-IR abilities.

My thinking has been, 'if Dan can get business with a rc based system, I should be able to do something with my c150'. I've looked at it and can see how to temporarily mount a vertical camera without an FAA STC. Assuming I can get coordinates for subject fields into my GPS before flight I could cover quite a few in a flight. But then I look around at what services and capabilities are currently available for agriculture and it's not encouraging. That's where my previous comment comes from. I guess I'd just want to conduct a trial-run to see how it goes.

[skymind]

Dan, Do you save and download your GPS trails for use in GeoMarking your pics?

[camship]

Oops

[kd7ost]

Dan, Do you save and download your GPS trails for use in GeoMarking your pics?

No, I wouldn't find that too practical especially since I'm flying circles in the area within visual range. The GPS trail would look a mess. I simply geo reference one or more locations that show up in the photo. Typically a corner or two of the field I'm shooting.

Remember, I'm flying this AP like an RC plane within close range of the field. The autonomous function is just in case and is only employed to bring the plane back to my location in the event of interference. Something I have never had happen in the 4 to 5 years I've been doing this. Don't confuse what I'm doing in agriculture with autonomous UAV activity. This is basic RCAP with safety features.

Dan

[skymind]

Thanks. My trails are tangled string also. Geotagging the exif is something I want to do and like Gene I was going to start to write software to merge my trail data with exif data to geostamp my pics. Not only for verticals but it would be helpful for my oblique hand-held shots.

But now just before I ordered an interface cable for my Lowrance and got going on the software I found the world of gps data loggers and software. So after looking these things over for a couple hours I'll go with Robogeo software and 'Holux M-241 Bluetooth Data Logger GPS' for its MTK chip, WAAS, display, inclusion of altitude recording, bluetooth, and yellow. Looks sporty. :)

[kd7ost]

How do you use something like that in our application?
Wouldn't a standard WAAS enabled Garmin GPS give the same information?

Dan

[skymind]

As for how it's used, I'm not sure how much detail you're looking.

To review, the concept goes like this. In pre-flight you sync the camera clock with the gps. (for some you do it manually, I saw where one gps can do it for you thru a cable) In flight the camera records the date/time with each image in exif data and the gps records (into gps memory) the lat/lon and timestamp of the trail for each point in the trail. (and gps altitude for the unit I'm buying) And then in post-flight processing, software takes the timestamp of each image to find the best matching timestamp from the gps trail log and retrieves lat/lon from that trail log point. Then it adds this lat/lon into the image's exif data.

What good is it? I'm having trouble due to the large number of pics I can take. The problem is sometimes I lose track of where these places are and really have to study Mapquest and Google maps to find them again. Geotagging and mapping software can really help.

Of course for oblique shots it's telling me where the airplane was but it would still help.

For vertical shots the gps tag would be more accurate to what you've photographed.

Various mapping tools can show you a map with little tags where pics were taken. Interfacing with Google Earth etc can be done.

Other software can then put the lat/lon and timestamp on the visible image if you want. Hopefully I can have some of this work done in batch mode to lessen my effort.

As for Garmin, I did read in gps forums where guys said they used Garmins for this but when looking at Garmin product specs I really didn't see it discussed. Nor did I see discussions on Garmin software elsewhere. I wanted to know that the software is available and have a good idea of what it does.

Hope it helps

[kd7ost]

I understand the concept I guess.

My thinking is that in a UAV type, autonomous flight where pictures are being commanded along the way that it would be pretty viable. Gene needs it for example when they're flying patterns and taking hundreds or more pictures in a search and rescue operation. As they pour through the pictures later, and spot something that needs to be looked at by a ground crew, the system gives them a Lat and Lon to send the ground crew to.

My Ag work is quite mundane by comparison. I take a picture, sometimes just a single shot, of a field that I'm parked by. Most of the time I'm in an edge of the picture. All I have to do is get the coordinates of one corner of the field and the GIS guys line up the software based on that single point. It would be very intensive to use the software process by comparison.

Dan

[skymind]

Looking around recently I saw that old conversation but since you asked I thought perhaps you shelved any effort there and forgot about it. In any case it sounds as if you get your needs dealt with.

Tom

[kd7ost]

Yes that was a question I had asked some time ago. But I have many projects I'm working on. I keep the Ag process separated from the other projects so when you asked it in this thread I thought it pertained to the Ag work. I see no real need for it in this type of AP.

Dan

[Totu]

Dan somewhere you mentioned using Telemaster-like planes now. Have you switched?

I'm also curious about competition you get will full-scale operations. Do you know of many fullscale guys who do this? While passing by the Caldwell airport a few weeks ago I noticed a banner on one hanger for aerial photography.

[kd7ost]

Dan somewhere you mentioned using Telemaster-like planes now. Have you switched?

I'm also curious about competition you get will full-scale operations. Do you know of many fullscale guys who do this? While passing by the Caldwell airport a few weeks ago I noticed a banner on one hanger for aerial photography.


I just use my twin boom pusher. I don't prefer tractor configurations and tail draggers with narrow LG when doing this. Telemaster would be a suitable choice of planes but it has shortcomings like all planes do. My approach to design and build my own is rather time consuming though. I started with a Rascal 110 and a G26 on front. I was glad to put a more solid plane on the ground.

I've never had a full scale operation compete with me or vice versa that I'm aware of. There are millions and millions of acres of crops in the U.S. alone and so far their is enough work for everyone.

Dan

[Brio3]

Hi Dan,

Andrew and I just finished a 1/4 scale Balsa USA cub with a G-20. It has the following istalled: Co-pilot / Picopilot, wireless downlink, OSD, etc..
Anyway, we maidened it last night and dialed in everything. The systems worked perfect and we took several photos. When we loaded them onto the computer they were all blurry :mad: . I mounted the motor with 3/4 inch rubber mounts but apparently the shock absorption is insufficient?
Not sure what to do? I'm thinking I need to convert it to electric?
Thanks for any input you can provide.
Hope all is well,

Steve Morris (the jet guys)

[kd7ost]

Hi Steve,

Haven't seen you in a while. Hope all is well. I miss seeing your son fly that jet. I haven't been out to the lake bed in months. I'm currently in California on fires down here and can't wait for a break to get home a few days.

I struggled with the vibration in my gas plane for a while too. I have some pretty good soft mounts on the engine but had the same issue with blurry pictures. I finally solved it with a camera mount that was weighted. Now the pictures are all fine. Describing the mount is tough and I don't have any pictures with me to post. It might be best to just show you what I did when I get home. Electric is good but so costly. Most of the time I think it will work out OK but on the marathon days I appreciate that I can just refuel and go back up. I can fly for many hours a day without having to recharge anything just by topping off the gas tank. It's your call though. It's not usual to have to take 15 to 20 different fields a day although I've done that a few times. Most of the time it's a 1 or 2 field deal and only takes about 10 to 20 minutes per flight.

Dan