How far do you think the recently announced US Drone Integration Pilot Program will improve our understanding of what the different roles will be for local authorities and air navigation service providers in a future UAS traffic management system?
Richard Parker I think we have to take a broad view of this and try to understand what people think is going to happen by having a UTM in place. This announcement, while very positive for the industry, is really about understanding what we really need from a UTM. It’s not about saying “tomorrow we will need this, this and this to enable this type of operation” but working out what is the interplay between local authorities, states, civil and national regulations and the entities in between. We have to work out what it is they’re trying to solve.
Today in the UK, for example, permissions required to fly in controlled airspace will need an interface with NATS and that will take place typically manually, through a paper- and voice-based process. In some areas where that permission is required NATS will have to work with other stakeholders. In London that might involve many different groups. A positive step from a UTM perspective would be to enable a digital workflow between all of those stakeholders – and the drone operator – so you can apply for and receive that permission digitally.
But is that UTM? For us today, UTM isn’t about what people think it will be in ten, 20, 30 years’ time with millions of autonomous drones flying around. UTM right now is simply about improving the communication in an increasingly automated fashion between all the various stakeholders involved in drone operations. And, as always, that comes down to the issue of enforcement. That’s the main political challenge in front of us. The reason people want rules and regulations is that they need to know that everything happening around them is happening safely.
Industry is pushing back on that because most people think it’s quite safe now. They think they should be able to go off and test their technologies in more scenarios. But it has to be done incrementally. Once we digitise the existing processes then the next step, for example, will be to bring in local law enforcement – police services who will need to quickly determine whether the drone currently operating nearby can be identified.
Some companies think that’s what needs to happen right now. They think we will need a big, interconnected – certainly national but maybe even international – system where all drones have telemetry flowing in real time. I think that that’s not potentially the best solution. It doesn’t really matter what drones are flying where; it only really matters if drone operations are taking place in a sensitive place – around a nuclear power station, an airport or a populated area.
We need to look more “needs-based” UTM where you say “with this type of operation, within the vicinity of these types of structures, then your drone has to meet certain capabilities.”
We can then start defining capability levels. Capability Level Zero may simply be you just need to have a mobile phone in your pocket so the airspace manager can call you in the event of an incident. If you want to go fully beyond the visual line of sight and carry cargo and you want to fly a drone that weighs over 150kg you may need the highest level of capability where you are visible to radar. So you’ve got to really address this whole complex landscape of current challenges before we can figure out what UTM means to all of the individuals who might be involved.
Even with one entity managing the airspace today it’s still an incredibly disconnected process and very human-centric. It’s difficult to see how the Drone Integration Pilot Program will change the nature of the engagement between the drone operator and the UTM system, today, and tomorrow the drone itself. As you go towards a more automated future the level of information you’re going to require in two directions is going to increase massively. So UTM for Altitude Angel is a vision and a journey and it comes with a set of expectations which are not about everything being connected all the time.
But don’t we need have to have some view of where we’re going to be in 10 years’ time, what sort of identification system for the drone, the owner and the pilot, for example?
Richard Parker I’m not saying we shouldn’t have big visions. We have a 35 year vision. We know where we’re going and what technologies will be required. But to say that you’ve got to have an endgame product now doesn’t make sense. Having a big vision means that you are trying to understand what is likely to become a requirement and that you’re sensitive to the needs of various different groups, and that you’re working to create technical solutions to address these issues.
So this is a bottom-up rather than a top-down approach? Or should the two have to work closely in parallel?
Richard Parker It’s a parallel requirement but there’s a degree of calibration that will need to happen. Technical companies need to understand that regulators are not there to impose boundaries but to enforce safety. And regulators need to understand that safety is not an excuse for slow progress, that safety should be an opportunity. And that’s what drones are, an opportunity. And we’re at risk globally of not being able to capitalise on that opportunity if we don’t cooperate.
It’s great to see the FAA announcement involves cooperation but that doesn’t mean an ANSP cooperating with a tech company it means a complete digitisation of that process. Before we can do anything in an automated fashion – much less improve what we currently do today – we need to just be able to answer some basic questions such as who is using the airspace right now? Can I identify the owner of that vehicle?
As we go forward we will need to know not just who is flying but what qualifications they have. We will not see an influx of drones into a particular area unless we also see an equal rise in the capability of relevant authorities to manage traffic there. And just because we think that something is the best solution, it doesn’t necessarily mean that it is. We should focus our technologies on foundation components and then create lighthouse scenarios along the way. We will need drone, pilot and operator registration systems and then tie those into our data-provision systems so the more we know about a person and their drone mission then all of our safety data becomes automatically more tailored.
For many, the dream of UTM is to remove humans from the loop. But for regulators their vision is fixed on beyond visual line of sight where a pilot is somewhere else. It doesn’t mean there is no pilot. So UTM should be a series of demonstrative steps to show regulators that industry can do these things in an increasingly autonomous way without compromising safety.
I think there will need to be more organisations involved in the UTM chain than there are in the aviation chain. Environmental protection agencies, for example, will want to know what drones will be flying in their airspace. So I think UTM also includes a constitutional component.
It’s great that people want to register drones but that’s just one millionth of the overall picture. We are going to need good data and for this local, state, national – and potentially even international regulators – don’t necessarily have to be in sync. We need a UTM which is flexible enough to allow local variations, to prove capabilities locally. And this is a difficult problem to tackle from a national perspective on day one.
Who’s going to pay for this evolutionary UTM system? The drone operators?
Richard Parker There are many models and we don’t favour one over another but there are clearly some more sensible approaches than others. I think, from a safety perspective, you want to encourage people to do the right thing. If you bring in a law that says: “You must register all your drones,” then you need to offer more of a carrot than a stick. Successful registration campaigns should include an element of education and quantification – so the regulator can see what’s happening.
One of my personal views is that the burden of running the system should be met – in the main – from those who stand to gain the most from it, such as commercial aerial users. That view might be opposed by some organisations but from a technology perspective it doesn’t matter to the process of UTM whether we charge or don’t charge. The cost for running a registration system should be fairly minimal and it should be one of the essential ingredients of a future UTM system. Unlike a lot of the registration systems today it shouldn’t be a standalone database. We want to encourage hundreds of drone companies to start creating great flight-planning software, automation software, data-reading software. But everything that enables that drone ecosystem to flourish requires an identity and the next step is what is to determine what that person is entitled to do which is different from everyone else.
I would register as a person into, for example, the Altitude Angel registration system and that would generate a profile. I would define a set of conditions that would have to be met by a commercial pilot, including a series of identity checks using data owned by the pilot but accessible by the state. We think it is really important that to participate in a UTM you don’t have to give up your privacy.
How should the industry work collectively to prevent accidents – and respond if and when an accident happens?
Richard Parker We have the Global UTM Association which seeks to be that collaborative voice. But like all trade bodies there’s a motive there to support the needs of its members and not necessarily the needs of the wider industry. I think it’s good that the industry is pulling together in certain areas. We have the Drone Alliance in Europe and the Drone Manufacturing Alliance but do we need one voice? Probably not. There is only a handful of very large drone manufacturers today and they’ve all got proprietary systems. But should we care about regulating for a technical standard or regulate for a requirement and let industry figure out what the best standard is? Competition will generally produce the best results.
But we do have more or less integrated safety standards on both sides of the Atlantic. There needs to be some agreement on where a drone should be to fly over a city. We need minimum standards and someone needs to say what they are.
Richard Parker That makes a lot of sense. But you have to answer the basic question: when does a drone stop being a toy and become an aircraft? And that’s a regulatory question, not a technical one.
Regulators have to agree on a series of capabilities that they expect drones to have and comply with before they can do particular types of work. And that’s what’s missing today. And very simply that would answer your airworthiness question. Do we care, for example, that a 2kg recreational drone hasn’t passed an airworthiness certification test if it’s flying above a big empty field?
In the hands of a bad pilot, a well-built aircraft is still a very dangerous prospect. UTM doesn’t make the safety problem go away because if you don’t want to participate in a UTM you’re in a different world and that means you will need counter-UAS systems to deal with uncooperative drones. Then you get into risk management and that, I think, is where the real regulations are required at the moment.
Five years ahead, what’s your company going to be doing that it’s not doing at the moment?
Richard Parker Looking at what the Europeans are planning with U-Space and what NASA and the FAA are doing with their UTM programme we would like to provide more of what we would term “navigation services”. Today we focus the bulk of our services on humans flying drones, providing services to connect that person with transient information about the area they’re flying in. Our Airspace Alerts service looks at where you’re going to fly and then connect you with information on nearby aircraft and works out if you’re likely to conflict. As we look further out, we’re definitely going to see us shift into more real-time services – including potentially command and control – of drones.
We’d also like to engage in very, very serious trials with autonomous, no-humans-in-the-loop flights. We’ve already built with our partners – including Imperial College London – 4D time separation and de-confliction services that we’ve tested over millions of iterations of simulated drone flights.
To whom will you be selling these services?
Richard Parker It depends on the timeframe. Today, most of our customers are B2B, governments and ANSPs. As a UTM company we’ve built technologies which enable us to work with aeronautical and environmental data to a degree that is not usually common in the aviation industry, and we’re finding that those technologies have a lot of appeal to existing aviation stakeholders before we even talk about drones.
Innovative ANSPs and other aviation companies are starting to wonder what their relevance is going to be going forwards. So they’re starting to procure from us UTM technologies which today prominently focus on safety applications, designed to target the pilot. Why are they doing that? They’re not licensed to do that in many respects. But they’re responsible for looking after the airspace. So if they can be shown to de-risk that airspace through their activities they can reduce the costs of their insurance policies, effectively paying for this work by reducing the risk. One of our biggest markets is in insurance.
Are you making money at the moment?
Richard Parker Yes. We’re very pleased to be making money. We’re about a year ahead of our schedule in terms of making significant revenue.
Many smaller ANSPs are still considering what steps they should take to develop an effective UTM system. What should be the first things they should do?
Richard Parker We’ve developed a “safety campaign in a box” service. It’s not just technology, it’s a service that comes with consultancy to help a country roll out an effective UTM system. As far as I’m aware, we’re still the only company that’s actually successfully done this on a national scale with the official sanction of the Civil Aviation Authority. What we’re trying to do is make it simple, affordable and effective for an ANSP, anywhere in the world, to say to its citizens: “We’ve digitised all the information that you will need to plan your flights safely and we’ve made it accessible in formats that you will understand.”
We’ve established a channel where drone pilots can start to interface with the ANSP so they can report where they’re flying and when, without requiring any really complex technology but with a software interface which will automatically update with whatever standards are introduced.
Which is why we think it’s so important to be working not just with the ANSP but also with manufacturers who need better data and support – and to work with the regulators who need to understand what’s available from a technical perspective.
If I were an ANSP planning to introduce a UTM system what should be my key priorities?
Richard Parker ANSPs regardless of size need to have a long-term plan. They have a right to manage the airspace in their country but it’s clear that if you play the UTM game out to its obvious conclusion then more air traffic management is going to become automated and that means UTM companies starting today are getting a head-start in delivering those services. In the future it may well be possible that ATM goes away and will be handled by a UTM company.
To secure their relevance in that world ANSPs need to start working with UTM companies. If they don’t there are potential revenue opportunities that they’re going to miss and they could become irrelevant players in an evolving UTM scene.
This is not good news for controllers
Richard Parker Look at automotive manufacturing – an industry that has become very automated over the years. What you typically find is that production workers have taken up other work in the factory. An air traffic controller today can only do so much with the combination of processes, rules and technology available to them. We have teams of controllers from different ANSPs we work with regularly who input into our development process to help us understand how we can – appropriately – enhance that experience. In our view, air traffic controllers are very switched-on individuals who can do things that most people can’t and we need to make sure they don’t have UTM foisted on them from above, but that they’re involved in designing that change from the ground-up.
We talk about ATM and UTM and the potential merging of the two but we’re absolutely adamant that as you move towards a more digital future you have to bring together different stakeholders. We can’t successfully automate unless we work with controllers to understand what kind of stuff we can automate and what stuff will get in the way of this.
Richard Parker is founder and CEO of Altitude Angel, a company which offers cloud-powered safety, analytics and advanced capabilities for unmanned aerial vehicles for commercial entities and consumers.