We were recently interviewed and asked to talk about how we got into UAVs and what tips we could share with others who are interested in doing so. It came out pretty well.
I am not a farmer but I’ve been living among them for years now. Some of my best friends are farmers, truckers, and mechanics and conversations naturally turn to farming and crops. I’ve also been getting heavily involved in precision agriculture through UAV/drone imagery and data collection so looking at crops and taking pictures of them is becoming more a part of my daily life.
Rather than produce imagery for a farmer today, I thought I’d produce some imagery for my friends to help communicate what “too much rain” looks like to a farmer.
That is soy, or should be soy, in the foreground. It is currently approximating a marsh. Corn is on the other side of the road and you can start to see the scale of the impact, and why getting drainage right is so important.
Again, soy in the foreground but less flooded. Some pretty healthy corn and then more corn impacted by the water. This photograph also demonstrates the value of aerial imagery – the field in the distance is not visible from ground level without driving through the standing corn closer in.
Last week we took a look at the FAA Section 333 exemption grant data to see what states had the most activity, how long the FAA took to issue grants, and what purposes were given for flying a UAV commercially. This week we want to explore what vendors and UAV models are most popular with commercial UAV operators.
Keep in mind that this data only covers organizations that chose to file a 333 exemption. Also, there is some lack of clarity in the models. For example, the Skywalker Flying Wing appears in the data. I believe that this represents someone doing a custom build using the Skywalker as a base.
We’ll answer the following questions:
- What is the most popular UAV model used by commercial operators with a 333 exemption?
- For vendors that have more than one model, which model is most popular?
- Where are those vendors located?
If you have additional questions that might be answered by analysis of this data set, please drop us a line and we’ll try to get it answered. Also, if anyone is interested in funding additional research, we’d love to collect and analyze similar data sets from other countries.
So what is the most popular UAV model used by commercial operators with a 333 exemption? The DJI Phantom 2V+.
DJI holds the top five spots and eight of the top ten spots. (This chart is truncated, the full chart is available at the end of this post.)
Next, for vendors that have more than one model, which model is most popular?
DJI’s most popular model is the Phantom Vision 2+, Sensefly’s eBee is their most popular model, and 3DR Iris+ leads their line. An example of the type of follow on question that comes out of this analysis is this – Why is the Altavian Nova F6500 more popular than their R8400?
Finally, where are those vendors located?
As we have been hearing for months, DJI dominates the US market for UAVs. But US vendors out number all other vendors combined. I wonder what that will look like in a year’s time? Two year’s time? I do not think that the market can sustain 50+ UAV vendors.
What impact will regulation have on these figures? What vendors will be able to successfully integrate transponders, for example, into their product line.
As operators start generating revenue, will they continue to use consumer grade UAVs or will they move to purpose built UAVs with more professional features?
Watching the vendor space mature and settle out will be very interesting.
Full chart of models listed in Section 333 grants:
There is a lot of uncertainty in the US UAV market due to lack of data. We do not have accurate sales data, number of hours flown, number of incidents, or even a good definition of what a UAV incident is. However, there is one really good UAV related dataset: Authorizations Granted Via Section 333 Exemptions. This is the FAA’s list of exemptions granted to allow commercial UAV operations.
Let’s ask some simple questions of the data, leaving the best for last. (If you don’t want to read the whole thing, please do pay attention to the last question.)
- How many petitions/exemptions by state?
- What was elapsed time between filing of petition and granting of exemption?
- How many petitions were there for each industry/purpose?
Before going forward please understand this – this post is not in any way a comment on the legality or appropriateness of the 333 process and it does not in any way imply that commercial operations without a 333 are in some way illegal or “worse” than operations with a 333 exemption.
Most counter-UAV techniques are illegal or very closely regulated outside of a war zone. The proliferation of consumer and commercial UAVs is prompting people to consider ways to disable them or take them down outside war zones. So far, we have seen shotguns, brooms and even kangaroos. How about some more precise options?
What follows is a thought exercise. Jamming, firing weapons indiscriminately, and even taking down a UAV with a net are all likely to be illegal wherever you might be reading this.
Most UAVs are controlled by a pilot using a radio transmitter or by a ground control system that instructs the UAV to fly to a particular set of waypoints. It is possible to jam the control signal, monitor the data channel, and even hijack the UAV. (A subject for another post.) This activity is certainly illegal.
A simple physical approach:
Acquiring and targeting a small UAV in motion with the Mark 1 human eyeball is tough. Attempting to shoot it down with a normal firearm is harder still. Doing so would violate several fundamentals of firearm safety:
- Always keep the muzzle pointed in a safe direction
- Be sure of your target and what’s beyond it.
If you really want to physically take down a UAV using a projectile weapon, consider using a net gun or a paintball rifle. (More on that later.)
Let us put aside “how to bring down the UAV” for a moment and address UAV detection. There are commercial solutions in this space and interested parties should explore those options for definitive, professional solutions. For the sake of this exercise, lets consider something like the Bluetooth Sniper Rifle.
Many consumer UAVs use 2.4GHz for command & control or data links. Such a “rifle” should detect a UAV using 2.4GHz at long range. Mount such a device on a tripod with a gimbal driven by a system that can point the detector in the direction of maximum signal strength. (Exercise left to the reader, as my professors used to say.) This provides a bearing and elevation from your location to the device you are targeting. You’ll have to spray the air space with rounds as you don’t have range information to provide a precise targeting solution but you could have the paintball gun fire in a pattern to place shots around the target. A few paintballs in the rotors should do the trick.
Now, tie two or more of these targeting systems together and you’ll have bearing, elevation, and range to the target. If you know the ballistics of your paintball rifle, you could probably place some pretty precise shots rather than spraying the air. It has been awhile since I used a paintball rifle but I recall that figuring out the trajectory of a paintball round was quite difficult.
[If anyone wants to try building this, please let me know and I’ll help.]
TrackingPoint – a precision solution:
TrackingPoint has a commercial solution for calculating targeting solutions for rifles, what they call “precision guided firearms”. Their solution uses a human to acquire and mark the target and then fire the round. Once the human marks a target, a computer tied into the rifle calculates the appropriate solution and guides the human to make the precise shot.
So the system takes targeting information in electronic form and uses it to provide a targeting solution. Can’t we take the targeting data from something similar to what we postulated above and use it to guide the human? I imagine so. (I asked TrackingPoint about this possibility and received no response.)
Keep the human on the trigger and use frangible, rubber, or other non-lethal rounds. You may have a long distance, precise, counter-UAV solution.
I will be presenting on forensic analysis processes and techniques at the following conferences this summer:
To whet your appetite for the presentation, here is an outline of the talk. You’ll note that we cover everything from why you should be interested in this material to an overview of a typical UAV to an analysis process to specific analysis steps for each component to real time analysis techniques.
Update: 5 days later. Frontier still can’t tell me anything, but they asked me to check with Mediacom. Mediacom agrees, Frontier is dropping the traffic, it is Frontier’s problem.
Hey David, Thanks for contacting us via email! I'm sorry to hear about the issues you're encountering when trying to access Frontier.com. Looking over the trace route since it's dropping on Frontier's end there is not much we can do from our side. Since you've already submitted a ticket with Frontier the best option would be to wait until they can resolve this issue. If you have any other issues or questions feel free to contact us again here! Thanks, Chris
Update: 12 hours later, Frontier still has no explanation for their apparent blocking of Mediacom IP addresses:
Ask Frontier @AskFrontier @dckovar Hi David, I wanted to let you know that we are still investigating this for resolution. -KH
I’ve been trying to pay my Frontier phone bill for a week now. Sure, I could use a check, but this is the age of the Internet and I’d also like to set up automatic billing.
But I cannot get to http://www.frontier.com. For days.
It finally dawns on me to do a traceroute:
traceroute to frontier.com (22.214.171.124), 64 hops max, 52 byte packets 1 router.asus.com (192.168.1.1) 2.671 ms 0.996 ms 1.035 ms 2 * * * 3 * * * 4 172.30.89.97 (172.30.89.97) 119.509 ms 12.377 ms 16.368 ms 5 172.30.35.69 (172.30.35.69) 29.432 ms 172.30.9.161 (172.30.9.161) 19.789 ms 14.916 ms 6 126.96.36.199 (188.8.131.52) 25.615 ms 25.803 ms 25.520 ms 7 cr2.kc9mo.ip.att.net (184.108.40.206) 33.190 ms 39.948 ms 31.791 ms 8 cr2.sl9mo.ip.att.net (220.127.116.11) 31.831 ms 30.923 ms 35.987 ms 9 cr2.cgcil.ip.att.net (18.104.22.168) 35.811 ms 32.269 ms 31.755 ms 10 ggr2.cgcil.ip.att.net (22.214.171.124) 96.169 ms 173.662 ms 31.426 ms 11 126.96.36.199 (188.8.131.52) 29.736 ms 29.467 ms 28.955 ms 12 ae2---0.cor01.chcg.il.frontiernet.net (184.108.40.206) 31.577 ms 33.271 ms 38.361 ms 13 ae1---0.car01.ftwy.in.frontiernet.net (220.127.116.11) 44.077 ms 46.896 ms 44.600 ms 14 * * * 15 * * * 16 * * * 17 * *^C
That is via a Mediacom connection. I switched over to a Frontier connection that I happen to have handy and ran the traceroute again:
traceroute to frontier.com (18.104.22.168), 64 hops max, 52 byte packets 1 192.168.1.1 (192.168.1.1) 3.033 ms 2.051 ms 1.182 ms 2 50-109-64-1.bltn.il.frontiernet.net (22.214.171.124) 25.556 ms 27.117 ms 24.808 ms 3 ae01---0.car02.bltn.il.frontiernet.net (126.96.36.199) 23.762 ms 35.961 ms 23.907 ms 4 ae3---0.car01.bltn.il.frontiernet.net (188.8.131.52) 38.245 ms 37.857 ms 36.997 ms 5 ae7---0.cor01.chcg.il.frontiernet.net (184.108.40.206) 91.912 ms 37.640 ms 27.502 ms 6 ae1---0.car01.ftwy.in.frontiernet.net (220.127.116.11) 115.455 ms 38.203 ms 38.660 ms 7 18.104.22.168 (22.214.171.124) 36.852 ms 41.143 ms 36.730 ms 8 126.96.36.199 (188.8.131.52) 38.250 ms 41.218 ms 37.705 ms 9 184.108.40.206 (220.127.116.11) 42.477 ms 37.212 ms 38.697 ms 10 * * * 11 * * * 12 ftr.com (18.104.22.168) 38.683 ms 38.508 ms 38.329 ms
It seems that 0.car01.ftwy.in.frontiernet.net does not accept traffic from a Mediacom source.
I’m sure this is legal, but it certainly makes me rethink using Frontier for any services.