I think we just learned something pretty important about the whole DJI situation here in the United States. And depending on how you look at it, it might raise even more questions about how much information we’re actually getting from our government when it comes to Chinese-made drones. Before I get into this, I want to give full credit to Haye over at DroneXL. He published an article that I think every drone pilot should read, because he obtained a Department of Homeland Security document that was not part of the public report. And what is in that document is pretty incredible.
DHS Flew Blue UAS Drones Through Hudson Yards
Here’s what happened. The Department of Homeland Security, through its National Urban Security Technology Laboratory, has been testing Blue UAS drones for first responders. Those are the drones that are allowed to be operated by government agencies in the United States. They meet certain federal sourcing and cybersecurity requirements and are supposed to give agencies alternatives to foreign-made drones like DJI.
One of these tests took place at Hudson Yards in New York City. That is a very brutal environment for a drone. You’ve got huge buildings, GPS interference, reflected signals, RF congestion, all of the things that can make flying a drone in a dense urban environment pretty difficult.
They tested several Blue UAS drones there. But here’s where it gets a little more interesting. They also tested two DJI drones, the Matrice 4T and the Matrice 30T. DHS actually scored those DJI drones right alongside the approved Blue UAS aircraft.
The Public Report Left the DJI Numbers Out
When the public DHS report came out, the DJI scores weren’t in it. The public report acknowledged that covered-nation drones were tested and said there were significant differences between them and the Blue UAS platforms, but it didn’t tell the general public what those differences were, or even which direction they went. Instead, it said the covered-nation results were available to public safety agencies upon request.
Well, now we know what those numbers were.
The DJI Matrice 4T Scored 4.6 Out of 5
According to the DHS addendum obtained by DroneXL, the DJI Matrice 4T scored a 4.6 out of 5 and the Matrice 30T scored a 4.3. The Skydio X10D scored a 3.7. And the Parrot Anafi USA scored just a 2.4.
It wasn’t just the overall score. According to the addendum, the Matrice 4T scored a 4.9 for its regular camera, 4.9 for its thermal camera, and 4.9 for command and control link quality.
Remember, this is Hudson Yards. This isn’t somebody flying over an empty farm field in North Dakota. This is about as challenging an RF and GPS environment as you’re going to find anywhere. And the DJI drones performed extremely well.
One Matrice 30T Drifted Into a Building
There’s another part of this that we absolutely have to mention, because I don’t want to cherry-pick the information that supports one side of the argument. One of the Matrice 30Ts had a problem. According to the report, during a nighttime flight in a narrow urban canyon, it lost both satellite positioning and vision positioning. It dropped into attitude mode and drifted into a structure. It was damaged badly enough that it couldn’t continue the test.
That’s pretty significant. If we’re going to talk about how well these drones performed, we also have to talk about when one of them fails. But even with that incident, DHS evaluators still gave the Matrice 30T an overall score of 4.3.
The Disadvantages DHS Listed Had Nothing to Do With Flying
Something else really caught my attention. According to the DroneXL article, the disadvantages DHS listed for the DJI aircraft included cybersecurity concerns associated with being manufactured in a covered nation, and the fact that buying these aircraft benefits a covered nation rather than domestic manufacturers.
Think about that for a second, because it gets directly into something a lot of you have been saying for years. How much of this is actually about cybersecurity, and how much of it is about building and protecting an American drone industry? Those aren’t necessarily the same thing.
This Test Doesn’t Settle the Security Question
I want to be very careful here, because this particular DHS assessment does not answer the cybersecurity question. In fact, the assessment specifically says it examined operator experience and did not evaluate or determine national security risk. So somebody can’t take these results and say, “See, this proves that DJI drones aren’t a security risk.” It doesn’t prove that.
But the opposite is also important. If we’re going to restrict these products because they’re supposedly an unacceptable security risk, then I think it’s completely reasonable for Americans to ask where the testing is that demonstrates that risk. And at the same time, when the government conducts an operational comparison showing how these drones actually perform against the alternatives, why weren’t all of those results included in the public report?
Maybe DHS has a legitimate reason. DroneXL reached out to them, and they said certain NUSTL findings are designated For Official Use Only because they contain sensitive operational information. Okay. But according to DroneXL, public safety agencies could already request these results from DHS. So I think it’s fair to ask why the scores themselves couldn’t have been included in the public report. I think we all know why.
The Cheaper Drone Won
There’s another number I think you’re going to find interesting. According to the assessment information, the Matrice 4T cost DHS about $12,958 and the Matrice 30T about $18,608. I’m not sure why they had to pay that much, because you can go on B&H Photo right now and get a 4T for about eight grand, but the government likes to pay more for stuff. The Skydio X10D was $28,382.
So put yourself in the position of a fire department or a police department or any kind of search and rescue organization. You’re not debating geopolitics. You’ve got somebody missing, or you’ve got a structure fire, or you’ve got a police situation unfolding. You need a drone that launches quickly, maintains connection, gives you a great thermal image, and actually works between buildings. And according to these test results, the much less expensive DJI Matrice 4T scored higher overall than the Blue UAS alternatives that were scored at Hudson Yards.
That’s not my opinion. Those are the scores contained in the DHS addendum. What is my opinion is that this information should have been part of the public conversation.
Test the Drones. All of Them.
Here’s what really bothers me about the entire DJI debate. For years, we’ve heard about the potential security threat posed by Chinese drones. We’ve heard about data security, supply chains, foreign adversaries, and protecting critical infrastructure. Those are legitimate things for our government to investigate.
But I’ve said this a thousand times and I’m going to continue to say it. Test the drones. Develop a cybersecurity standard. Make DJI pass it, but also make Skydio pass it. Make Parrot pass it. Make every drone manufacturer pass the exact same test, and if DJI fails, show us the results. Then we actually have evidence to discuss.
What I don’t like is making enormous decisions about an entire industry while relevant information sits somewhere that the people affected by those decisions can’t easily see.
This addendum doesn’t prove that DJI has no cybersecurity problems. It doesn’t prove that every American-made drone is inferior. And it doesn’t prove that every restriction on Chinese drones is purely protectionism. But it does tell us something we didn’t know before. Our own government tested DJI drones against some of the alternatives that agencies are being encouraged, or in some cases required, to use. And according to this addendum, the DJI drones performed extremely well. Yet when the public version of the report came out, those scores weren’t available. I think we’re allowed to ask why.
Where Do You Land on This?
Does this change your opinion at all about what’s happening with DJI in the United States? Do you think these results should have been included in the public report? And I think the bigger question is this: if the government believes DJI represents a national security threat, should it have to publicly show the technical evidence behind that conclusion? Or are there legitimate national security reasons why some of that information simply can’t be made public? Go read Haye’s original DroneXL article for yourself, because there’s a lot more in there than I covered in the video.
This article is based on a video from our YouTube partner Russ. Watch the full video on the 51 Drones YouTube channel, and find more of Russ’s work on his DroneXL author page.
