Last December, Kittyhawk, Wing and AirMap all traveled to San Bruno, California, and demonstrated an implementation of an open source, network-based, remote identification solution called InterUSS. Designed to find the right balance between privacy and transparency, they not only established that Remote ID for drones is possible, but that it can be enabled without any special hardware in an exceedingly simple manner.
Kittyhawk founder Joshua Ziering has already answered the basic questions around what was demonstrated in San Bruno and why Kittyhawk favors network based Remote ID. We were able to connect with him to explore these topics and more in order to discover not only what the ultimate Remote ID solution should look like, but what it will mean to see it enabled.
Jeremiah Karpowicz: Exactly what Remote ID for drones should look like has been a major topic of discussion of late, and there are ongoing arguments about what the requirements of such a system should be, how much information should be made available in it, who should have access to all of that data, etc. Ultimately, are all these conversations and discussions more about technology, culture or policy?
Joshua Ziering: All of this comes down to a single question that’s being asked by stakeholders throughout the space, and especially by members of the public. That question is, “who’s operating that drone?” And that’s really the core issue we’re facing today.
When someone sees a drone in the air, their question usually isn’t, “what if something happens?” but rather, “what are they doing?” That’s a fundamentally different question. Being able to identify a drone from a remote location is going to be paramount because that’s going to allow the public to feel comfortable when they see a drone in the air. They’ll either already know or can easily find out exactly what it’s doing.
Ultimately, all of this exists at the intersection of technology, culture and policy, so it’s really not about any one of those things. That’s just one of the reasons why we think that a network-based solution using smartphones is the best and most approachable way for the majority of operators to fulfill a Remote ID requirement.
There are lots of hobbyists out there who don’t feel any kind of identification requirement should apply to them. I’ve seen it explained in terms of the license plate metaphor, because while we do require cars that drive on the road to have license plates, we don’t require bikes that are on the road to have those same license plates.
I don’t think that’s a strong metaphor, and here’s why:
Have you ever seen a bicycle take down a tractor-trailer? It’s apples and oranges because the reality is that we’re not talking about the same actors in the same kind of situation. A hobbyist might be doing something very different in the airspace than someone using a drone to inspect a roof, but you can’t change physics just because you put a different word on it.
I certainly understand anyone who’s concerned that what we’re creating with Remote ID is going to be an overly onerous solution. If something costs a lot of money or is difficult to use, then it’s really not providing the most possible value for the public. What you’re looking at with a dedicated, mandated hardware solution or something less accessible to the public is a potentially costly solution that doesn’t ultimately solve the problem, and that’s not what anyone wants.
I think we all want something that’s readily approachable, easy to deploy, low cost, and provides a ton of value to the largest segment of the public. We wanted to find out how we could solve these issues efficiently and at scale. That’s why we think InterUSS is the right way to solve this problem.
Sally French laid out the essentials associated with your InterUSS demonstration, which is a system that provides everyone with the ability to easily identify who’s flying or responsible for a drone flying nearby. What would you say are the biggest benefits of this kind of system?
I think there are two things. The transparency it enables is big, but the privacy aspect of the system is just as important to highlight.
One of the elements that ultimately attracted us to the InterUSS platform and solution is that it walks a very nuanced line between showing people what they need to know, but not making too much information available. Someone can’t peer into the entire national airspace and see where everyone’s flying all at once, or the history of what all those flights looked like. This is an ephemeral solution that gives information as it’s needed, and when it’s no longer needed that information goes away. There’s not a blockchain style record of flights that are happening. This is what you need to know, when you need to know, given to who needs to know.
It’s not any more onerous than that, and that helps protect operator privacy as well. For example, you wouldn’t want a company to have to disclose to an outside party that they’re delivering something to your home. UPS doesn’t have that mandate. If you see a UPS truck, you don’t know which specific houses or locations it’s going to stop to make deliveries. At the same time, if you’re concerned about where a drone is flying or what it’s doing, this gives you the information you need to have those concerns allayed.
You explored in detail not only why an open-source approach is critical for Remote ID, but also why open-source is plenty secure. Some people hear the word “open” and think we’re talking about something that’s inherently insecure, but that really isn’t the case, is it?
I find that programmers tend to be the largest contingent of people who understand how secure open source actually is, and that’s because they know what it means to build, test and use systems and networks like the ones we’re talking about.
When you look at your bank or all the different security protocols that we’ve all agreed on as best practices in the security industry, almost every single one of them is open source. After all, how can you have security unless you have as many eyes as possible looking at that code? If it’s just a black box that has someone saying, “this is secure”, that invites scrutiny from every angle. More often than not, the scrutiny is justified because it discovers that the emperor isn’t wearing any clothes, and that vulnerability can be exploited if the party that maintains that code is not willing to fix it in a timely manner.
Open source software allows companies to take their destiny into their own hands, fix things they see as broken or insecure, and then upstream those changes back into the larger project for the benefit of the whole community. That’s part of the reason an open source, network-based solution for Remote ID is going to benefit the drone industry as a whole in multiple ways.
Can you talk a little bit about the difference between the network-based Remote ID that Kittyhawk favors versus the one that companies like DJI prefer?
We are not actually opposed to DJI’s approach. We think there’s room for both. Depending on the application, we think different types of remote ID have their place.
If you’re simply flying a line of sight mission, and it’s going to be in a limited area, we think that a broadcast based remote ID solution is one possible way you could comply with federal regulation that could possibly be enacted in the future. If you’re doing a BVLOS operation, and you’re doing something like inspecting a pipeline, a broadcast based remote ID solution starts to have some problems.
The main one is that you have to be in range of the broadcasting so that if a drone flies over your house and two minutes later you want to know what that was, you no longer have the ability to reach the aircraft. Moreover, in this situation, things like data sharing become very difficult. The network-based remote ID solution not only makes it easy for operators to share their information, but it also makes it easy for concerned stakeholders and law enforcement officers to find out what they need to know.
We think there’s room for both solutions, but ultimately, for larger operations that are more complex, network-based remote ID is going to be the way of the future.
What would you say to concerns that what you’re talking about is inherently too onerous, or that the system you’re advocating for is making too much information available?
I really don’t think we’re asking too much. Part of the appeal of an open sourced solution is that everybody can see what’s happening. As contributions are made and features are rolled out, there’s going to be fewer and fewer surprises and issues.
Secondly, most of the people that we work with are actually eager to share their information. They don’t want to bother law enforcement and deal with getting interrogated on the side of the street while they’re trying to do their job. They can choose what they want to share, and who they want to share it with. We think that’s a key difference between us and other platforms. We want to give the operators as much discretion as possible when it comes to sharing information about what they’re doing and who they’re doing it for.
Is it too much? Will the number of companies that are involved make it all too complicated? I don’t think so, because it’s going to be paramount for this industry to work together, and with our InterUSS demonstration, we proved what’s possible when that happens.
In light of the fact that the InterUSS demonstration had you closely working with companies that are nominally your competitors, how do you think this sort of collaboration will impact where things go with Remote ID? All of the companies that participated in the demonstration have different initiatives around what they want to in the airspace and within the industry, don’t they?
It’s also important to mention that there were three different types of aircraft involved in this, not all made by the same manufacturer. Not all running the same software. Really, the only common element amongst them was the InterUSS platform. Even though each company has very different mission statements, and at some points we even compete with one another, we were able to come together and agree on this set of standards that are going to give us a singular result that creates Remote ID in the national airspace. I think it’s a real shining star example that if we can do it, the whole industry can do it.
We don’t need a walled garden approach here. Collaboration is going to be key to moving all of this forward. If you look at the full-scale aviation world, there’s actually some precedent here.
There’s something called MEL, which stands for minimum equipment list. Depending on the type of aircraft you’re flying, and the airspace that you’re flying in, you are required to carry different equipment. So if you’re within The mode C veil, you’re going to need to have ADS-B identification. But if you’re not, you won’t actually need to have ADS-B on your aircraft.
In a similar fashion, we could say the drone industry could take a note from this. If you’re just flying a basic mission to take pictures of a bridge or you just want to inspect a roof, maybe a broadcast-based remote ID solution will provide key stakeholders the info they need to know. But if you want to do more complex operations, a network-based remote ID solution is really going to be the answer.
What will it mean for the InterUSS solution to define what Remote ID will look like for the drone industry as a whole?
One of the advantages in using Kittyhawk is that we’ve been ready to deploy a remote ID solution since yesterday, partly because of the same issue that all of our enterprise customers are facing: the first people to show up to any given operation is always law enforcement. Our customers have to answer the same questions, for the same people, over and over. Despite the fact that they’re licensed, insured and completely authorized to be there, they go through the same song and dance every time. So all of our existing customers will be able to check that box that they’re now compliant with Remote ID, and we can offer that service right now.
As we start to see a rollout of more and more regulation that will get rid of waivers, we’re going to see Remote ID as a base layer. Depending on what you’re doing, whether it’s BVLOS delivery or pipeline inspection, if you are not in immediate eyesight of your drone, you’re going to need to implement a remote ID solution. I think that’s a common-sense way for accountability to be brought to an operator who isn’t anywhere near the aircraft.