#62 Andrew Hately, Eurocontrol: U-Space, UTM, and the integrated airspace

Welcome back to The Vertical Space and a conversation with Andrew Hately - a senior researcher in the Drone unit at the EUROCONTROL Innovation Hub in France - about U-Space, UTM, and ATM. 

Andrew emphasizes the necessity of radical steps to integrate the airspace for UAS traffic management, suggesting that without such integration, UAS operations may not be economically viable. He particularly focuses on U-Space, discussing its origins, current status, requirements, and the challenges in its implementation. Andrew advocates for a proactive approach in designing the U-Space system from the outset, rather than adopting a slow iterative process, to ensure its effectiveness and eventual success. He also addresses the importance of drone detection, categorizing potential threats as careless, clueless, or criminal, and discusses the challenges in detecting noncooperative aircraft.

Moreover, Andrew delves into the various technologies used for drone detection, including radar, acoustic, optical, and radio-based systems, highlighting the need for both tactical and strategic deconfliction to ensure safety in airspace shared by manned and unmanned aircraft. He acknowledges the current lack of confidence in surveillance systems for strategic maneuvering and emphasizes the necessity of knowing the precise locations of aircraft for effective tactical maneuvers. Additionally, Andrew discusses the role of commercial service providers in UAS traffic management, recognizing the challenges they face in a nascent market with limited flying activity but expressing optimism for the future of UAS operations. Overall, Andrew's insights reflect a well-informed and balanced perspective on the evolving landscape of UAS and UTM.

Transcript

Andrew: 0:26

We wanted to see from the beginning where we wanted to get to at the end. Although it will be an iterative process, we didn't want it to be a process in which each step was going to be the most convenient step to take right now, without a view of where the path would lead. And we wanted to see clearly what's the goal, and then work back how do we get there. And if the goal as I see it, is that we end up more or less integrating UTM and current air traffic control then we need to accomplish a number of steps, which are going to be quite difficult. And so I have the tendency to draw attention to these difficult steps right now, when people say we might do these later, or hopefully maybe do these never because they're difficult. And I, I feel if we take that approach we will have never reached the goal and the benefit that we really would like. So I tend to argue against the expedient and the convenient, and I'm pushing towards difficult steps and so for some of my colleagues, these steps really are tough.

Jim: 1:44

Hey, welcome back to"The Vertical Space" and a conversation with Andrew Hately, a senior researcher in the Drone unit at the EUROCONTROL Innovation Hub in France. Andrew talks about what major steps must be taken with UAS traffic management and U-Space, and that radical steps have to be taken to integrate the airspace. And if we're going to get the high density operations we need an airspace where all aircraft are going to fly together. And if we don't do it, UAS is unlikely to be economically viable. He extensively discusses U-Space. How it started. Where it stands today, the requirements for setting up U-Space, why it's taken so long to implement and how it'll evolve in the future. Something really revealing about Andrew, in my opinion, is his comment early in our discussion, as he says, we wanted to see from the beginning where we wanted to get to in the end. Where in other parts of the world, UTM is being deployed in a slow iterative process, perhaps the most convenient steps, but not the best steps. Andrew wants to design the system from the beginning and the way he envisions it, looking in the end. And that the expedient and the convenient, we'll never get to where we want to eventually be. That was a really good discussion. Another good section is on drone detection. He refers to the categories of the careless, the clueless and the criminal, and how detection has to account for each. He discusses radar acoustic optical radio-based systems and others, and the challenges of detecting noncooperative aircraft. He mentioned that detect and avoid is not a focus of U-Space today. And how U-Space plans do safely separate vehicles. And to that end, we have a great discussion around tactical and strategic deconfliction. He explains that so much as strategic today because he believes they currently don't have any confidence in this surveillance. And then in order to do tactical maneuvers, you need to know where the aircraft are. Listen how Andrew believes manned and unmanned aircraft will be managed in the same traffic management system. And what will become an automatic traffic management system. Yes I agree with Andrew this is certainly where people tend to get a bit anxious, including some who've been on our podcast. We admire his bold albeit controversial approach to the future of traffic management. He talks a lot about commercial organizations or service providers who are providing or plan to provide traffic management functions. Who they are and where they're based and their commercial and business opportunities. He discusses that these are very tough times to be in the service provider business. Because there's not enough flying yet. He says we suffered from a bootstrap problem in this domain that until there's traffic management there won't be flying and until there's flying, there won't be traffic management. So somebody has to get it started and these companies are all hoping it starts soon. It was an absolute delight to speak with Andrew. He's so informed, he's been doing this for a long time, he's extraordinarily articulate and he has a realistic, balanced, yet optimistic view of how UAS and UTM operations will come together and has some clear areas where UAS operations have real and unique value today. Andrew, thanks for joining us and to our listeners, we hope you enjoy our conversation with Andrew Hatley as you innovate and The Vertical Space. Andrew Hatley is a senior researcher in the Drone unit at the EUROCONTROL Innovation Hub in France. He was the technical coordinator of the CORUS and CORUS-XUAM projects, which produce updated additions of the concept of operations for U-Space. Andrew applies the results of research in supporting EASA’s generation of acceptable means of compliance material in standardization groups and is a member of the ICAO Advanced Air Mobility Study Group. Andrew has worked in research at EUROCONTROL for more than a decade. Prior to research Andrew developed flight plan processing software in EUROCONTROL’s Network Management directorate. Andrew Hately, welcome to The Vertical Space. Great having you on.

Andrew: 6:17

Thank you very much. I'm very happy to be here.

Jim: 6:20

First question, is there anything that very few in the industry agree with you on?

Andrew: 6:24

Yeah, there, there is. I am from a research background, and I have been looking at this, UTM, U-Space stuff, and I've been thinking from the beginning, where do we want to go eventually? And I see that there are some radical steps that we need to take. And I have had some disagreements with some people, even some quite notable people on Is it worth taking these steps? Are these steps going to be too costly or too difficult? But I would say that if we are going to get to high density of operations, we're going to get to, electric vehicles carrying passengers, then we need to have, an airspace where all of the aircraft can fly together. And that's going to take a revision of flight rules, which might influence the airspace categorization classes, and I think these steps are a bit scary for some people. Some people look at the current situation and say, how do we get there? So, there are a few people in the industry who disagree with me on taking radical steps, and are these steps justified? But I would say if you look at the alternative, we will just never get to the situation where UAS are economically viable if we don't take these steps.

Luka: 7:41

Is the disagreement due to the fact that people don't necessarily see the additional value prop or is the disagreement in terms of the complexity, both technical and organizational, to pull something like that off?

Andrew: 7:53

So the disagreement is mostly about how do we make the step. For example, would it involve retraining people or would it involve modifying aircraft and things like that? And there's always this resistance because the initial impact will be so large. And so that, that's why people tend to disagree with me, but I would always argue that not doing this, the impact will be even larger. So at a certain point we have to grasp the nettle and take a difficult decision. So. I know I recently had a disagreement with Maria from EASA about whether or not we could have a UTM system, a U space system, ultimately controlling manned aircraft. And she felt that was perhaps going to be such a difficult step that there wouldn't be support for it in the industry and so on.

Luka: 8:46

Maybe it's important to highlight the difference in the definitions between the U. S. and, Europe. Because I'm guessing in the context that you're referring to, UTM is, increasing automation in managing air traffic, as opposed to traffic management of unmanned aerial systems. Is that correct?

Andrew: 9:08

Yeah. So when we started, the term UTM was being used as a general term. Now, the specific implementation in America has quite a defined scope and is a lot less general than the way that we were using the term in the research community. The European version, USPACE, we started to look at that from a research point of view as being ultimately something that would go on to control these eVTOLs and similar things like that. UTM as it has been implemented in the USA is very much about supporting current operations, supporting VLOS flights, and allowing VLOS flights to occur in an airspace where there are possibly VFR flights going on and things like that. I will tend to use the term UTM more in the general sense, meaning any traffic management of unmanned aerial systems. And typically we are looking at this being completely unmanned traffic management or, as unmanned as we can make it, there will always be a need for escalation to a human for obscure corner cases. We've seen this in many forms of automation, but the general operation we expect will be automatic and it will be providing a service. Because it's automatic, it'll be providing a service which is scalable to a very large amount of traffic, and it'll be providing that service very cheaply. So the scope of USPACE is much broader than the current term UTM is used in the U. S., which is really focusing on the current operations of VLOS flights in the most part. I would imagine as, this part 108, is published for consultation in the summer. it may be that the term UTM begins to take on a broader meaning in the USA. Because I think part 108 will push the envelope a bit and push it towards some of the decisions which have already been made in Europe. do you want me to contrast the USPACE and UTM? Would that be helpful?

Jim: 11:08

Great.

Andrew: 11:09

Yeah. So in, in Europe, USPACE is something which is proposed by the regulator and the USPACE service provider has certain obligations towards the regulator. In the USA, the UTM system, all of the responsibility rests with the aircraft operator and the UTM service provider is a subcontractor of the aircraft operator, but all of the liability still rests with the aircraft operator at all times. So the aircraft operator is seeking support, but that's all he gets from the UTM service provider. Whereas in USPACE, the USPACE service provider is something a little bit like an air navigation service provider in that they have their own responsibilities and they are meeting these towards the regulator rather than only towards the aircraft operator.

Jim: 12:05

Andrew, if you could, I think there's subtleties to how people disagree with you. Explain a little bit more about, why you are not at the same sheet. So for example, so many people may say there's going to be an iterative approach to UTM or to U space where you may be feeling, I want to go further faster. Others may be also saying, and I'm now I'm cheating a little bit because we talked a little bit about this in advance, you see the ATC piece being integrated much faster into UTM than others would say, and that's clearly getting into a controversial area. So could you dispel it a little bit more?

Andrew: 12:45

I think it comes from, how we started. the group I was working in started looking at, UTM and USPACE. We wanted to see from the beginning where we wanted to get to at the end. Although it will be an iterative process, we didn't want it to be a process in which each step was going to be the most convenient step to take right now, without a view of where the path would lead. And we wanted to see clearly what's the goal, and then work back how do we get there. And if the goal as I see it, is that we end up more or less integrating UTM and current air traffic control then we need to accomplish a number of steps, which are going to be quite difficult. And so I have the tendency to draw attention to these difficult steps right now, when people say we might do these later, or hopefully maybe do these never because they're difficult. And I, I feel if we take that approach. We will have never reached the goal and the benefit that we really would like. So I tend to argue against the expedient and the convenient. And I'm pushing towards difficult steps and for some of these, so for some of my colleagues, these steps really are tough. If we want to qualify, an automatic system as being an ANSP, then we need to learn a lot about how we qualify these systems, what the expectations are. Today, if you look at, an Air Navigation Service Provider, we certify these with reference to the responsibilities of people, with reference to the responsibilities of management, and it's much harder to take that and say, well, how could we implement all of this stuff as software? What's the algorithmic basis on which you could provide these services without a human there? And it, I think there's a lot of worries in the minds of people who have to actually practically do this, which Perhaps they're not pressing on me so hard as I'm in the research domain. it's much easier for me to be a visionary because, well, I won't be alone in trying to implement this for one thing,

Jim: 14:58

So you're also advocating, listening carefully. There are people who are beating up their phones right now, I think, listening to what you just said. So you're also advocating some form of automated air traffic control here, which would have less human involvement. So I would assume this is another controversial part of what you're saying, but you may not be saying that, so please explain in more detail.

Andrew: 15:21

I think that's where we get to, but it doesn't start like that. So it starts with, we want to fly UAS in the lower airspace. And we would like to do this as cheaply as possible. So we would like to provide, traffic management services mostly on the basis of software. Probably with a human supervisor for these unusual corner cases for escalation in case of an emergency. But for the general running, Software based. Once we have a software based traffic management system, we have the question about what do we do when a manned aircraft wanders in there for whatever reason. If it's a manned aircraft which is expected, for example some helicopter emergency medical service flight, or a manned aircraft which is unexpected, some light aircraft which has had an engine failure, how do we deal with that? And the current approach which What is proposed is to deal with this as an exception, and we will have the UAS move out of the way and the manned aircraft will be treated as an emergency case. But in the longer term, if we're going to do this in a scalable manner, that manned aircraft has to be treated as a routine process. So eventually, we would like the manned aircraft to participate in the traffic management system, which is providing a service for the UAS. So the, I would like that the manned aircraft is filing a plan, and then is being given an authorization, and then is being monitored that it's following its plan. And once we start to do that, we see that we have a system which can be scaled upwards. These, electrically powered, taxi flights, these eVTOLs, passenger carrying operations, which many operators would like to do without a pilot on board, will probably be managed by a similar, UAS traffic management system. In Europe, we expect USPACE to manage these, but they'll be flying a bit higher than the bottom level, 500 feet. They'll be flying up to a few thousand feet. So now we have a situation where this is an airspace where manned aircraft might fly more than just in case of emergency. And again, I would propose that the logical way to do this in a scalable manner would be to integrate the manned and the unmanned aircraft so that they're controlled by the same system, and so that they are able to interact in a safe but integrated way, and that requires that we have an automatic system which is now giving instructions to manned aircraft, not as an exception, and then we need a qualified automatic system, which is capable of safety of life operations and is good enough to do this. I think that's where people begin to get a bit anxious and begin to disagree with me. I want to go too fast. I'm not sure if I do want to go too fast, but I have a clear view of where I want to get to.

Luka: 18:17

So Andrew, if this is what success looks like ultimately with UTM, can you go in a bit more detail to describe the milestones that will lead to this success? Both on the, technical side, implementation, cultural, organizational. what are all the stakeholders who have a say in this progression?

Andrew: 18:38

Okay, so there's quite a few people involved there. So, The first is that we need to be confident that we can do it. So we need the industry, which is producing the systems, to be able to demonstrate that these systems are good enough and have a high enough quality and a high enough, availability and resilience that they are capable of being trusted to do this. The second step is we need to understand how it will be done. And one of the key problems here, which I would like to spend more time working on, is the flight rule. We do not today fly small UAS following either visual flight rules or instrument flight rules. And if you look at the way that ICAO describes airspace classes, they're all described in terms of these two flight rules. So what we need to do is identify a flight rule for the small UAS, and then understand how it can safely interoperate with the existing ones or the existing manned traffic. One option would be that there's a new flight rule that all of the traffic follows, both manned and unmanned. Another option is that we would find a third flight rule which is compatible with the other two in some way. If you look at the descriptions of how, flights are supposed to occur, it is possible to mix visual flight rules and instrument flight rules today. If we had a third flight rule, perhaps we could find a way to mix that. So, we need to understand the operational basis on which the traffic is mixed together. We need to understand the technical basis on which it's done. What are the procedures? Are the systems capable of providing them? We need to convince the regulator that this is going to be safe enough. It should meet the target level of safety. and we need to convince the existing aviation community this approach is going to be acceptable to them. That they are going to certainly not be inconvenienced and hopefully will benefit from this approach. And we're a little bit optimistic on this basis because, studies have shown that current visual flight rules are not really as safe as people believe they might be, and particularly, tests about visual acquisition of the target show that it's quite common that a pilot of a VFR flight doesn't see everything that they, should. So we are hoping that by providing some support, which is being originated to allow, remotely operated aircraft to fly, that we can enhance the experience of pilot onboard aircraft and that way we can get acceptance for these kinds of what would otherwise be really fairly radical changes to the way that the airspace is structured.

Luka: 21:27

What is the primary benefit of increasing automation in the air traffic management function? Is there mostly a safety argument? Would you say that this is the primary problem that the industry is facing, or is it more about productivity, being able to accommodate greater traffic density, because part of it, and I'd love your thoughts on it as well, the way traffic management is done in Europe is not the same, obviously, as it's done in the U S organizationally, but at the same time, it feels like there is more inefficiencies and please correct me if I'm wrong in Europe because of the fragmented market, and every nation wanting to have that function internally, as opposed to perhaps sharing it among multiple countries. A couple of questions in there, and I know there's a lot to unpack there, but, what comes to mind?

Andrew: 22:17

Okay. So there, there are several points there, which I'll need to unpack. Yes. The European air traffic management system is quite fragmented and complicated because air traffic management is the, responsibility of sovereign nations. And in the European Union, we have 27. In, in Eurocontrol, we have 41 member states. So each of these is able to operate its airspace as it wishes, but we work together, and you probably know that Europe has a single integrated traffic management system operated by Eurocontrol, the network manager, and we in the network manager, we ensure that each air traffic control center, well, ensure, we aim that each air traffic control center does not suffer from more traffic than it's able to manage. So there is a process, a rolling process every day where the air traffic control centers look at what kinds of effects might influence their ability to manage traffic, weather effects. I mean, weather is a dominant phenomenon in American air traffic management. it's more prevalent here only in the height of summer and winter. But still weather takes a role. You also have, more subtle effects like important football matches or other things like that which cause, increases in the flow of traffic. All of this is discussed every day in a rolling program and then in the network management system we implement delayed takeoff to avoid that there is, too much traffic in the air, more traffic than can be safely managed. So this goes back to the start of the question where you said, is this a, an issue of safety or is it an issue of capacity? Safety is never compromised, but the result is that when we hit a hard capacity limit, the demand cannot be accommodated at the time it would like to be. So, in Europe, we have these take off slots, which are issued for traffic management reasons, and that is quite evident. In the summer months in Europe, quite a number of sectors of the European sky are at the limit. The traffic cannot be increased without quite some evolution in how we manage the traffic. Certainly among the most busy airspaces in Europe is the region above Belgium, which is a crossing point for traffic coming off the Atlantic Ocean, heading to various European airports, and also traffic flying from Scandinavia towards Spain, and so it's a bit of a crossroads there. And in, in the Belgian sectors, the, Maastricht Air, Upper Air Traffic Control Center really is often pushing towards what it can deal with. And the result, unfortunately, is that the aircraft takeoffs have to be delayed. So, continuously in, in European ATM research, we're looking at how can we improve the capacity without compromising the safety. Many of these kinds of questions can be addressed by some of the techniques that we're looking at to use for UAS traffic management. One of the kind of the most basic principles is we've divided the air traffic management world into sectors. Each sector has normally a pair of air traffic controllers, one of them is talking on the radio to the aircraft, he's the executive, the other air traffic controller is coordinating with other sectors, and the limit to how much work they can do is the mental capacity of these two Air Traffic Controllers to think about the number of aircraft in their sector. For a sector where the traffic mostly fly in straight lines, that capacity can be very high. For a sector where the traffic have to be steered towards an airport or something like that, the capacity is much lower. But it's a mental limit. If the system is completely automated, a computer has no mental limit. We can just buy a faster and faster processor or scale it among multiple computers. So the limits become other than mental capacity. And that way we can, develop techniques in UAS traffic management, which will be re applicable in manned traffic management and promise to increase capacity and possibly reduce costs also. And those aims I think will be in the interest of all airspace users, all air traffic is looking for these kinds of benefits. So there's a great deal of interest from the air traffic management world in what we're doing in the UAS traffic management. But also, as a primary job, we would like to enable UAS traffic. The European Commission and the European, Parliament have identified that operation of small UAS will provide economic stimulus. There will be jobs resulting from these kinds of activities. So they're keen for this to happen as an end in itself. But the benefit that we can accrue from that, we can also apply to air traffic in general.

Peter: 27:39

And Andrew, in that vision that you lay out for the transition to automated handling of, that air traffic, is there agreement on the roadmap for how that information would be communicated to the aircraft themselves and how we move from where we are today to have a different way of communicating that information that its own pipeline is much, much higher capacity than voice over VHF that we have today?

Andrew: 28:04

Okay, we're at the beginning of this. So, we have a vision of this today. We are not even close to implementation. So the roadmap to get there will start by, let's understand the problem and solve it for the small UAS. And that's going to take us several years. it'll be 2050 or later, before we're beginning to apply this stuff to all upper air traffic. So in, In beginning to attempt to answer these questions, yes, we are looking at what's called, Controller Pilot Data Link, CPDLC, which is being used today in Europe for what you might call low priority communication between the air traffic controller and the pilot. So effectively, they're sending text messages to each other. The use of this I anticipate will increase, and in order to increase, it will require that we have low latency communications. When CPDLC started, there was a lot of worry about the latency, the messages might take a long time. But as we get used to using them, then yes, this could be a way forward. This could be a way to allow further automation of the air traffic control system and then conveying the, information to the aircraft in a more speedy and timely manner.

Peter: 29:20

Are people also looking at what that workflow looks like for pilots and controllers where you might have mixed usage of voice over VHF and then the rollout of a data based link and how do people who are in flow, who are, working the traffic, how do they bridge across both of those modalities? Are people, at the same time as they're discussing this roadmap, are they looking at that migration as well?

Andrew: 29:49

Well, that migration is ongoing. I've been involved in, simulation to do with researching this at least nearly 20 years ago, and the early use of CPDLC has been, well over 15 years. It is an ongoing activity and it's also used in the US and elsewhere. The crucial step is to trust it to be speedy enough to use for the kind of separation clearances. Today we're using it for the, lower speed stuff, the, uh, change the radio frequency and contact the other sector type instructions. But, for data link to be, the only means of communication, we need to get a bit more confidence about how well it works and how speedily it works.

Jim: 30:36

Andrew, let's talk a little bit about Air Risk Mitigation. Let's bring it back to drone air traffic management. How is EASA approaching Air Risk Mitigation?

Andrew: 30:46

So, today the U space regulations describe a thing called a U space airspace. So the state will declare a U-Space airspace, and that's where these services will be provided and where there will be used. So this process of declaring a U-Space airspace, it's a kind of little reserved area which is dedicated towards flying, UAS BVLOS flights, and so the state has to, or the competent authority as it's called, it's often the Civil Aviation Authority or the National Safety Agency, but it's an agency which is interested in air traffic, will have to look at the region where they want to create this thing, and they will have to create what they call a concept of operation, what they expect the UAS to do, how they expect them to fly, the processes they want to follow, the expected performance, what is the, navigation performance they expect from the aircraft, what speeds are expected, what's the latency that they expect. So once they have this concept of operations and they have to make a kind of estimate of the traffic, they will then do this airspace, risk assessment in which they will balance all of these factors and they will come up with a, A description of how this USPACE airspace can be operated safely or meeting the target level of safety which has been put forward for the area. So that is in the bubble of the USPACE airspace. So EASA today is segregating the UAS into these little regions. For a flight outside these regions, EASA has described an open category of flight, which is visual line of sight, which is done, in a manner where basically the UAS has to get out of the way when something is coming. And the other option is certified operations, which at the moment, there's not really much said by EASA. That would be something that would need to be worked out between the aircraft operator and the Civil Aviation Authority, probably with some kind of specific agreement about how it would be done. Maybe with a danger area created or, temporarily segregated area created for the flight to occur. So these are the, three means of operating today. And this is how, air risk is being handled. In the U-Space regulation, you find two other aspects which relate to air risk. There's this idea of dynamic airspace reconfiguration, and in this process, if necessary, U Space airspace will be temporarily closed to allow a manned aircraft to pass. And this is considered to be done when the airspace was previously controlled airspace, and this process would allow it to revert to being controlled airspace. So, air traffic control could decide to close the used space to allow an aircraft to fly through safely. The third method, is this electronic conspicuity where a manned aircraft could enter the U space airspace so long as it was visible to U space. And this is, in my personal opinion, the least well described. I could imagine this working as an exceptional process, but I'm not so happy with it being a regular thing, because effectively the manned aircraft comes along and all of the U-Space traffic has to get out of the way, and in doing so may encounter their own problems. So, this is the current approach to air risk. It has to be said that the current European, regulations anticipate quite low density of traffic, and we would be looking forward to a revision of the regulations, as the density of traffic is rising, and we start to see a need for more subtle, air traffic management for the UAS.

Luka: 34:51

Can we be a bit more specific and maybe run through a scenario of a drone operator who wants to fly a commercial drone delivery service or any commercial drone beyond visual line of sight in a, suburban slash rural area. And let's put the ground risk aside and just focus on the air risk. The challenge that exists today, and again, correct me if I'm wrong, but if one looks at the SORA air risk matrix, essentially any time that you're flying below 500 feet above ground, unless you are flying in uncontrolled airspace over rural areas, you pretty much immediately fall in, in the Air Risk Charlie class, which automatically classifies the operation as SAIL 4. The problem, though, is there's no tactical, TMPRs the tactical mitigations for the residual air risk that are approved for, medium and high integrity levels. So the challenge for these drone operators becomes; how can we somehow reduce the air risk class by either flying in U space, which is automatically class Bravo, or I shouldn't be using class Bravo because that confuses with traditional airspace, but ARC B, or they go and run an analysis of the actual air traffic, and show that the risk of encountering, manned traffic is actually lower than what the original SORA matrix would suggest, and therefore they get that class reduction. So talk about this challenge that operators today face. What are the levers that they usually pull on? Is this description correct?

Andrew: 36:38

Your description is perfect, and I'm very impressed with your knowledge of the domain. Yes, this is exactly what's going on. And if you look at, and I must say it's not something I'm particularly expert on, but if you look at the operations today, which are occurring, you have Manna operating in Ireland. You have Wing operating in Finland, you have soon Amazon operating Italy. They are typically having to deal with the fact that there's not yet uSpace available, but uSpace is coming soon. And as you correctly described, one of the things that they have to do is they have to make an argument about is the air risk acceptable? Is the risk of manned traffic coming by low enough? And typically that can be mitigated by asking for a restriction. and in Finland, you find, occasional restrictions are placed with the word Wing in the title, so you know who's operating there, but, either a restriction or, as you say, an analysis of the traffic, but that analysis, which you're talking about is the first small iteration towards this air risk assessment. Air risk assessment is a more thorough and long term version of this analysis. And so setting up a U-Space airspace is a logical extension of the process that, an exceptional operation would need to go through. So today, these operations are not in U-Space airspace, but the hope is that very soon they will be, we're expecting 2024 to see the first U-Space airspace is declared and it clearly makes sense to put them where people are currently wanting to fly. So, I would hope to see their emergence, in and around some of these areas where these operations are occurring.

Luka: 38:28

What are the requirements for setting up U-Space? Is there a threshold in terms of the traffic density that an area needs to have in order to warrant U-Space implementation?

Andrew: 38:40

No, there's nothing like that. It's a a choice to be made by the competent authority in each state. So we may see, U-Space airspaces which are set up in result of a demand from an Aircraft operator. We may see them being set up as a result of a demand from a service provider. We have seen a lot of interest in flying in harbors, for example, and there the harbor operator might be the one who would be, pushing for, a U-Space airspace to, to be declared. Or we might see states who want to set up U-Space airspaces because they have an expectation that once one exists that flying will begin. It's hard to predict at the moment because we're in a special period at the moment. Everybody is feeling their way and learning how to do this. I think once we have one or two U-Space airspaces declared, I think we will, then gain more experience about how to create them, more experience about how they can be used, and we'll see a period of fine tuning, but also a period when more and more are created as the states master the process and are more confident about declaring these things. And it must be said that there is quite a lot of work to do in the airspace risk assessment, because at a certain point there needs to be a link made between, for example, separation and the performance of the various systems, the performance of the aircraft, the performance of the traffic management system, the performance of the surveillance as it is. So, this is why the states are going quite slowly at the moment, but I think once it's been done a few times, I think the states will gain confidence, and it will become something which is done more readily. And then I would see in Europe that it would be, not unreasonable that a request from an aircraft operator, certainly if it's going to be more than just a few flights, might result in the work being done to, to set up a U-Space airspace.

Luka: 40:47

Talk more, Andrew, about why it's taken so long to implement U-Space.

Andrew: 40:52

It's really been a learning process. The U-Space regulation was published, in 2021, and then it took us quite a long time to develop the acceptable means of compliance. And I would say this initial acceptable means of compliance will probably be amended as we learn more. But it still has a number of gaps. There's still quite a lot to learn about, what standards to apply, how to make the implementation, how to certify the service providers. What the, Airspace Risk Assessment comes out at. What is the minimum required performance for the aircraft in order to achieve a separation which is fine enough to make the U-Space airspace have a capacity that makes it worth operating? We can achieve safety by restricting the flights a great deal, but that's not the aim of U-Space U-Space should allow BVLOS flight to be approved relatively quickly and often that the expectation is that the aircraft operator will put in a plan and generally that plan will be approved. We are hoping to avoid that, that planning results in rejection and conflict detection as a, a frequent occurrence. It should be a rare occurrence. We're hoping to promote the use of the UAS by, creating these U-Space airspaces. So the, airspace risk assessment should hopefully, produce a separation requirement which allows a reasonable capacity. And the state, or the competent authority which is making the airspace risk assessment needs to be convinced that the performance of the aircraft, the navigation performance, its ability to fly, the surveillance system, the satellite navigation and so on. All produce a situation where this separation can be achieved and produce a convincing story that it can be achieved repeatedly.

Jim: 42:55

Andrew, where and when do you see U-Space being deployed in Europe in what countries and why?

Andrew: 42:59

I have a reasonable expectation that we'll see something in 2024, probably Spain, Switzerland, and I'm hoping in Antwerp, there's a project called BIRDI, a European funded research project, which is trying to create, U-Space airspace over Antwerp harbour and I think they have a good chance. They, have a lot of resources and they're working hard towards this. But at the same time, there's a kind of unofficial race going on. I think Zurich are doing well and, I think we'll see one in Spain certainly in 2024, if at all possible.

Jim: 43:37

Drone Detection, how are they handling it in the U-Space? What are the different ways they're addressing it?

Andrew: 43:41

So drone detection has not been considered part of U-Space so far. And this is perhaps, it's worth noting that, security matters are the domain of the state. And, EASA, European Aviation Safety Authority, tends not to, get too much involved in saying things about what states should do. So, the U-Space regulation doesn't talk about drone detection, but we have a lot of interest in drone detection in Europe, and we're seeing quite a lot of activity, often driven by, state security organizations, often relating to airports, but other areas as well. And, the drone detection really is aiming at three different groups, referred to as the careless, the clueless and the criminal. And for the careless and the clueless, we can do quite a lot because we may require fit of equipment to the aircraft, which makes them detectable. The criminal are the third group and we're in an arms race with them. It's always going to be, what can we do and what can they avoid us doing? So, I think progress is being made in this field. Initially there was a lot of interest in primary radar, but that's very difficult. The aircraft presents a very small target for radar. The reflection is quite weak. The, small UAS move around at similar speeds to cars in some cases, or to birds. So it's easy to get, false positives from other things. So it's a difficult radar target, but there are methods that can be used. Other techniques which can be used include listening out for the command and control link radio signals. But again, in the case that somebody really doesn't want their drone to be observed, they can find ways around this. There has been some interesting work about drone detection in terms of one aircraft detecting another based on listening. There was a project called PercEvite, which was working on a hear and avoid, mechanism. And, that looked quite promising. And there's been a number of projects have been looking at see and avoid based on cameras and image recognition, sometimes stereoscopic, sometimes, with a single camera, these have in the kind of prototypes required quite a lot of electricity to be put into the, computing process to do the image recognition. But perhaps with a bit of refinement, these could become practical. It's not something that we're particularly concerned about in U-Space yet, because in U-Space we consider detect and avoid to be a safety net rather than the primary means of separation. So we are not at the moment requiring detect and avoid to be something fitted to aircraft in U-Space. Instead, we will separate the aircraft very widely so we feel that the residual risk is low enough that the detect and avoid is not necessitated. It will probably become something more significant when we go to higher risk operations, either because the density increases or because we start to have passenger carrying or things like that. So, there the risk might push us towards requiring, a safety net in terms of a collision avoidance system on the aircraft. And there, there is a lot of work being done. But at the moment, U-Space is not mandating any particular implementation.

Luka: 47:17

So in the U-Spaces that you expect to come online this year, the DAA function, as you say, will just be, assured in terms of making sure that the airplanes and drones never come close to each other or the drones themselves right. Is there a difference in how you are thinking about separating drones from manned aviation as opposed to drones versus drones?

Andrew: 47:45

So the current functioning of U-Space, the drones are separated by means of the planning. So any drone flying in U-Space must have a plan. It's called a flight authorization request. And these flight authorization requests are de conflicted. So the idea is that we de conflict the plans. If the plans are followed, then we've de conflicted the flights. We can't do this with manned aircraft that wander in. So instead we have this requirement that the manned aircraft are electronically conspicuous. The electronic conspicuity is handled by the U-Space service provider, so it becomes part of the traffic information which is shown to the pilot of the drone. Now the traffic information which is shown to the pilot of the drone differentiates between nearby aircraft which are conforming to their plan, which are, the pilot would be aware of but not concerned by. And nearby aircraft which are not conforming to their plan, which are potentially a threat. These manned aircraft, which are electronically conspicuous, will definitely be in the threat category. So, the, requirement in U-Space is that the drone takes appropriate action, which will be to get out of the way. So that means that this, handling of the manned aircraft, it, it should be safe, but it is going to be an exceptional process. It is going to require that the drone deviates from its plan. I await experience to see how this works, how well we can accommodate manned aircraft, I would not be very comfortable if this was a frequent occurrence, but certainly as it's been described, it is intended to be an infrequent occurrence.

Luka: 49:34

And in those cases where multiple drone operators are flying in the same area, how will drones be de conflicted and separated from each other?

Andrew: 49:42

That's done by this planning. The drones will all need to plan. The plans will all be de conflicted. Even if they're using different U-Space service providers, the U-Space service providers between themselves will arrange that the plans are all de conflicted.

Jim: 49:58

So these service providers are commercial organizations?

Andrew: 50:01

Yes. So, the, approach which has been, encapsulated in the U-Space regulation is that the provision of services is a competitive commercial activity. So, we expect multiple service providers, or we hope for multiple service providers and that they will compete, they will innovate, they will offer added value to their customers. The U-Space regulation describes six services which they should provide, four of which are always required. The other two may be required depending on what the, Competent Authority says about the U-Space airspace and we expect that these commercial operators will offer their customers a service which fits the customer's need. If the aircraft operator has a need for, high precision, high efficiency, they're probably willing to pay for that, and they will be paying for tools which allow the kind of things they want to do. If the aircraft operator has lower requirements, they may be able to get by with a cheaper offering. So we expect these U-Space service providers to occupy niches, have particular specializations and to innovate in the area. So EASA, and the aim of the European Union was to create hopefully a single European market for U-Space service provision. So we would see the emergence of U-Space service providers who could operate in multiple European countries. We would get economies of scale and we would quite quickly arrive at innovation high performance and, hopefully the lowest possible price.

Luka: 51:47

Andrew, this four dimensional flight path de confliction strategic planning issue, there's been some research that showed that you really can't be very confident looking more than, several minutes, into the future because things happen in real life. There's weather, there's contingencies, there's all kinds of things. How well do you think that this approach will, handle the real world?

Andrew: 52:15

So there's a continuous trade off. The, plan that the aircraft operator is asked to submit is not a line, but a series of boxes. And we asked the aircraft operator to plan these boxes, as large as they need to be that the aircraft operator has a confidence of 95 percent that they can follow this plan. So, it's not clearly spelled out, but if an aircraft operator is consistently planning too small, they will deviate very often and they will often end up in a contingency situation and they will have to invoke a contingency plan because they're unable to follow the plan that they've submitted. The other option is that the aircraft operator is very conservative and, submits enormous boxes. They say, like, all of the city for all of the afternoon, and they will simply find that they're in conflict with everybody else. So, there should be a natural pressure that the aircraft operator begins to master their uncertainties, that they begin to understand what weather will bring, and they will plan in function of their ability to fly. We, would anticipate that they would hopefully get better at this. We would hope that they could vary their speed, for example, they wouldn't always fly as fast as possible. So they have a bit of margin to deal with unexpected gusts of winds and so on. So we would hope that, in a mature U-Space, the aircraft operators are able to plan in a manner, which is sufficiently sized so that they can conform to the plan, but sufficiently small that they don't simply conflict with everybody else.

Peter: 53:59

Is it likely that their uncertainty will increase the more minutes into the future that they're planning? You know, they might be able to be quite precise four minutes, five minutes, seven minutes into the future in terms of where the aircraft is going to be in the plan, but if it was a very long flight, would that change towards the end of the flight or is that something that people are mitigating?

Andrew: 54:21

We, that, that model is generally true, but there has been work in manned aviation with what's called, trajectory based operations, where we can essentially have the idea of a kind of four dimensional contract where the aircraft will speed up and slow down in order to be within the agreed performance. The question is really a trade off. How much is the aircraft operator prepared to sacrifice in terms of flying at the most efficient speed and the straightest line compared with flying in order to be able to get their plan approved. So we would imagine there'll be a bit of give and take there, a bit of trade off, and I would imagine that once a U-Space airspace is running, the aircraft operator would be, well, complaining, would be reporting to the competent authority if the planning constraints were holding back their business. There is likely to be a requirement that the, competent authority would revisit the airspace risk assessment. And one input into that is the expected traffic. And I would expect that the aircraft operators would probably take the opportunity to express their desire for a different compromise on separation should they find it too constraining.

Peter: 55:49

Do people discuss things like, you know, afternoon thunderstorms that come up that might throw the entire plan out the window in this type of an architecture?

Andrew: 55:57

They do, but certainly, bad weather is probably less of an issue in Europe than in the US. We do get thunderstorms here, but it's probably not as dominant an effect on air traffic as it is in the USA.

Luka: 56:13

Andrew, is there a level of, flight density, operation density, at which point we really need to move away from this strategic 4D planning.

Andrew: 56:24

Absolutely. Yeah, absolutely. Now, the trigger, surprisingly, might not be the density, because, eventually, when we can't master the uncertainties at the strategic phase, we have to try and master the uncertainties at the tactical phase. And then we would be required to do tactical conflict resolution. Once we have tactical conflict resolution, we can allow more residual risk to remain after the strategic. But one of the big drivers of uncertainty are passengers. So I could quite imagine that the introduction of passenger operations, these eVTOLs carrying passengers around, would quite quickly push towards the need for tactical services. The passenger typically is going to have an impact on the take off time predictability. If the person has to get into the vehicle, might be their first flight, might feel a bit unsure, might feel a bit frightened or unwell. I could quite, imagine that relating to a, it takes longer than expected to close the door and then they miss the takeoff time. So I could quite imagine that we would have to go to a process where the aircraft are not de conflicted strategically, but are subject to a demand capacity balancing process and then are being de conflicted tactically. And I would say that's going to be the big push, but once we have Tactical Deconfliction, we can use it to increase the capacity of U-Space airspaces as needed.

Luka: 58:02

I mean, this entire conversation around strategic planning, just evokes the famous Mike Tyson quote that everybody has a plan until they get punched in the mouth. So, and, so, why especially among aviation professionals that know that things happen in the real world. It never goes according to plan. It's weather, it's, all kinds of different things, right? And so, why is there so little emphasis on tactical means is it because the technology is not there?

Andrew: 58:33

Okay. So why is it all strategic? Well, because we currently don't have any confidence in the surveillance. In order to do tactical maneuvers, you need to know where the aircraft are. You need to know with a known level of uncertainty where the aircraft are. And we don't today know what that level of uncertainty is. You will notice in the U-Space regulation that conformance monitoring is optional today. That's because conformance monitoring also requires that we have confidence in and have characterized the level of the surveillance that we can locate the aircraft during flight. And we don't today have that level of confidence. So the initial use of U-Space will be almost completely based on this strategic de confliction, and because of that, the margins for all of these uncertainties will have to be very large, so the aircraft will be very far apart. Once we begin to see flights, and we begin to get some experience, we can begin to characterize how well does the surveillance work. And in the U-Space regulation, there's this thing called Network Remote Identification, where the aircraft is sending to the U-Space service provider its position and the time and its identity. We will observe that for quite a while before we begin to rely on that. There may be opportunities to confirm how well that system is working. Interestingly enough, one of the users of U-Space is Israel, and in Israel, for security reasons, they have been doing a lot of monitoring the position of these aircraft by other means, and have been able to determine reasonably well how well this network remote ID is working. And certainly they have now scaled their separation minima in function of their confidence in this network remote ID. In the rest of the U-Space community, we still lack that experience, and I would expect most competent authorities would want to see traffic in their own cities, in their own U-Space airspace, and understand how well the surveillance is working before they begin to rely on it. So this is why we will have essentially have to have a time delay before we can get to any kind of tactical service. The first tactical service would be conformance monitoring. Conformance monitoring would allow, these margins for uncertainty to be reduced. They would allow the separation to be reduced a bit. And then, once we're really confident we would be able to introduce tactical services where we're giving instructions to de conflict the aircraft.

Luka: 1:01:17

What is the rough timeline for this?

Andrew: 1:01:19

I, optimistically, we'll see this stuff experimentally very short term but to have it rolled out generally, it could be another 10 years before everybody is really confident about this and it becomes routine, widespread.

Luka: 1:01:38

You mentioned a little bit, different ways of, detecting. Acoustic, Optical, et cetera. Can you spend a bit more time, and perhaps broaden that list, including Radar and some of the others and comment the pros and cons, of each and what you believe is the most attractive option.

Andrew: 1:01:58

To detect a non cooperative target, the problem with radar is that you illuminate the target with your own radio. So the amount of power that needs to be pushed out in order to get an echo back is enormous. The power falls off in proportion to the fourth power of the distance. So, you can be transmitting thousands of Watts and get microWatts back. So it's a very difficult job to do. And as I said, there are similar targets which will give clutter. There are cars and birds and things like that. Typically, we're seeing radar reserved for situations where the high cost, can be justified despite the low range and the not fantastic performance of these systems, but still it's better than nothing. Much more interesting to use optical where the illumination comes from the sun and you only have to capture the reflection. There is work going on there. I haven't been so involved in using optical, processes to detect drones for example, surveillance, but in terms of, collision avoidance, there's been quite a lot of research done in universities about this. And I think we're potentially seeing, this leading to practical systems in the reasonably short term. Hear and avoid is in the very early days, but it potentially, has quite some benefit. It's very good at detecting light aircraft, for example, or helicopters, which are especially noisy. Then we get onto collaborative means. If the other aircraft would like to be detected, radio based systems are obviously one option. The problem there, though, is finding radio frequencies which are available. Proposals for the electronic conspicuity which are coming from EASA include, one of these industrial scientific and medical radio frequencies, which is all very well, but it may suffer from interference. Another option would be, acoustic or, optical broadcasting, so, some kind of flashing light or ultrasound or something like that. And I think work is going on all of these, but, Yeah, I haven't been involved particularly in any research project which is looking at that. My, my main interest so far has been traffic management rather than on vehicle systems. Our colleagues over, at, Johns Hopkins are looking at a, there is a, some work going on developing a replacement for TCAS. TCAS in the generic term is referred to in the industry as ACAS, and the project which was developing a replacement for that was ACASX. ACASX spawned some variants, one of which was ACAS XU, which was for unmanned. Quite some work has been done on that. And, Josh Silbermann at Johns Hopkins was leading a project called ACAS SXU for small unmanned. And that was pretty promising. I've not been involved in that for a year or more, but it looked like it was going to go somewhere. So that was looking at using a system which would be compatible with manned aircraft. So, I think if that comes to, fruition, it's quite likely that standard would be adopted quite widely.

Peter: 1:05:22

Has there been any research on the potential for detecting the RF energy that's being emitted by the drone's propulsion system, perhaps the speed controllers or other sources of, what most of us would refer to as EMI, but detecting that in a useful way. Is there any potential there?

Andrew: 1:05:40

There is potential, but I've not been involved in any project which has looked at that. Projects which I've been aware of have more focused on the command and control link, which, is easier to detect and also easier to interrupt. There have been some, different approaches with that, because certainly, you can buy this box from DJI, this aeroscope to listen in on their communications. It makes it more obvious about what to jam if you want to jam the thing, and also you can listen on the telemetry. You can find out where the aircraft is, where the aircraft operator is, and things like that from listening into the command and control. So it's very beneficial to do that. The problem is if somebody doesn't want to be seen, they will encrypt that or they will use other methods. If they're already criminal in the way that they operate the aircraft, they can be criminal in the way they use the radio spectrum. So it is a harder task to detect and stop somebody who doesn't want to be detected and stopped.

Jim: 1:06:52

Let's talk a little bit more about the service organizations you've talked about, Andrew. What's the barrier to entry for some of these organizations.

Andrew: 1:07:00

The barrier to entry is probably, not enormous, but it's certainly there are fewer companies pitching into the service provision area than we saw emerge trying to build aircraft. So maybe it's not such fun. But if you look at the ones who are active today, I would single out the team at Wing as having put a lot of energy into developing standards and trying to create an environment, for U-Space for UTM. But Wing are a bit unusual in that they have no particular desire to be a service provider. They are, they see service provision as being just a necessary step so that they can become an aircraft operator. Amazon have also been active in that area. Again, Amazon, I believe, will only be their own service provider, but they've certainly contributed to the standards which are being evolved, and they've taken, quite some role in, in driving the industry. So those two are combined aircraft operator service providers. In terms of who else is in the service provision? You have a few companies who are pure service providers. I would, mention Dronic in Germany and, D Flight in Italy. and each of those kind of emerged before the U-Space regulation was published. And at that time, there was a slight expectation that they would become the kind of the appointed national service provider. But they are purely service providers. Behind them, there are companies writing software for service providers, and there are more of those, and I would mention. ANRA is one, which is operating in Europe and the USA. Unifly is based in Belgium. They've done well, supplying software to different, operating companies. OneSky, another, American company, which is seen in Europe. Drone Radar in Poland. and some of these, although they are primarily aiming at being software suppliers, some of them are in the process of becoming certified as service providers to prove themselves and also to make up for the fact that other service providers are not stepping forward. I know that ANRA are in the process of being certified at the moment, and we might see them offering services possibly in Switzerland, I don't know, but, they may see that as a necessary step. There are some of the larger ATM companies involved in the business, like Collins, Frequentis, Indra, and Thales, and all of these are looking at parts of the business, Frequentis is probably more looking at the other end. In the European model, there is the U-Space service provider, which, delivers services to the UAS operator. And then there's another body called The Common Information Service Provider. And Frequentis are looking at supplying software to those. The Common Information Service Provider is, a role which the state may appoint one or may decide not to and that role is involved in distributing information to everybody. And so Frequentis are looking at that, Indra are looking at that. So there are different larger companies looking at joining this market. I've met most of these companies through participation in European research projects. So when we've been, looking at, doing research, I've particularly been involved in research projects, looking at really fundamental questions at low maturity. I've worked with many of these companies, in these research projects, trying to answer very basic questions. How should things work? How do we solve these simple questions. What does a flight plan look like? And how do we deal with it? And so on. So, that's where I've met them. I've, I could mention even more. There's, Airbus UTM, who were previously known as Altiscope. They have a simulator. I'm working with them, they have a reference implementation of UTM in their simulator. There's a strong team there, they're very active in the development of standards, they're very active in the, work that's being done at ICAO, looking out how we can, integrate, this UAS stuff into the ICAO annexes and documents. We also have other smaller companies, the Technological Institute of Galicia has developed some UTM system, which has been adopted by Boeing. Boeing have a team in Madrid who are really expert at simulation of aerodynamics. And so Boeing have developed a simulator for, UAS which uses the ITG system to offer a complete package. There's a company called Open Skies operating. It's a very small company, but they've developed an open source, implementation of a kind of uTM system, which although it doesn't, isn't fully functional or it's partly functional, it is very compliant with the available standards. And because of the standards, compliance, their tool, Flight Blender, is widely being used for testing, often in combination with the, automatic testing mechanism of the, inter USS, which is an implementation of the, Discovery and Synchronization Service. Inter USS is also open source and it's a reference implementation of the DSS and includes within it a testing function. People are using that to test their system and they're using FlightBlender as a reference to compare their system against.

Luka: 1:12:54

The big outstanding question is how do these companies make money, right? What's the willingness to pay?

Andrew: 1:12:59

That's a question I simply can't answer. These are very tough times to be in this business, because there's not enough flying yet. We suffer a bootstrap problem in this domain that until there's traffic management, there won't be flying. Until there's flying, there won't be traffic management. So somebody has to get it started, and these companies are all hoping it starts soon. Because without flying, there won't be a revenue stream for many of them. I still wonder how many, the market has room for. I have good friends in all these companies. I wish them all very well, but honestly, it's a tough business. I don't envy their position.

Jim: 1:13:35

Andrew talk about the origin of U-Space.

Andrew: 1:13:38

So U-Space, we started with, an idea that there would be traffic management and in the research domain, we looked to the end goal from the beginning. And the end goal was that we would be able to control all traffic, including passenger carrying operations. The other thing that happened quite quickly in Europe was, the European Commission identified that UAS was going to be an area of important commercial activity. So, they wanted to promote U-Space. So, top down, Europe got to work on regulations. Regulations are valuable in Europe because we are 27 European Union member states. It would be quite likely that without a top down European regulation, we would have 27 completely different implementations. So the view of the European Union was that they would like to promote a single market. They wanted a harmonized environment where an operator, an aircraft operator or a U-Space service provider operating in one European country could move to a neighboring European country without difficulty. So this is why in Europe there was a rush to produce a U-Space regulation. This U-Space regulation has described how to fly beyond visual line of sight and it's described an airspace in which this is possible called a U-Space airspace. Now in America you have a totally different set of problems and you have a totally different environment. The Federal Aviation Authority is both the regulator and the service provider. And it was quite clear that Congress wasn't excited about the FAA becoming a service provider to drones. So I would say that the momentum in the USA was held back. It wasn't being driven by the FAA. Industry was driving this, wanted to do some stuff. The FAA was following on, making it possible, but observing, and we now see the gestation of this part 108. It will hopefully be available for comment this year. I think I slightly envy the position the FAA are in, in that they will be able to write Part 108, taking into account experience and observation of flying, which is, which has taken place, without their regulation. And so I think, the rest of us will look at the American regulation. I mean, Some people might say, Oh, it's behind, but. It won't be behind. I think we're all moving at the same speed. The American regulation will benefit from having been written, having taken into account the experience that's being gained, for example, at the moment in Fort Worth, with seven different operators flying, seven different service providers supporting them. And so the learnings from that experience can be fed back into regulation. Yes, there are big differences in the current situation, but I think in general, we're pretty similar in what's driving this, that the drones are there and we need to take care of them. They may pose a risk to the other aviation. They may pose a risk to the public. In general, the Aviation Safety Authority is tasked with dealing with that. So drone regulation is a natural outcome. And the second is that in, in both the US and in Europe, there is a political imperative to enable this stuff. So the regulation is not purely restrictive. It is trying to let this stuff occur. So I, I think, we are following slightly different paths, but we're going to the same place quite quickly. And, in my domain, in the research domain, we have a healthy cooperation with both NASA and the FAA. We meet, we compare notes, we discuss. So there is not any fundamental difference in outlook or opinion. But there are some. Some differences in just, like, the way things have ended up. Our regulation came first. Our regulation particularly pushed this idea that service provision would be a competitive market. It could be, that in some states they're not bothered about that. They might want to simply appoint a single service provider, as is often done for, manned aviation. So you know, that's a choice that Europe has made so it's not necessarily the only choice, but it's the one that, in which I'm working. And, it's a similar choice that has been made in the USA. FAA doesn't want to be the service provider, so there'll be a market for service provision.

Luka: 1:18:15

What notable differences exist in the way that the FAA and EASA think about strategic and tactical deconfliction? And especially when multiple operators are flying in the same area.

Andrew: 1:18:27

I think the, this is most visible when you look at what's being done in the USA to support the passenger operations. So the work that's been done in the USA, on the passenger carrying operations started with this question. Is it when, well, it started with the question, what is it? And then they might discover that it was similar to, managing small UAS. In Europe, we started with the question, can we extend the system, which is designed to manage small UAS? can we extend that system to manage passenger operations? So in America, you'll hear these terms like pSU, Provider of Services to, I don't know what the U is, UAM. and we have discovered from the work done at NASA that it's very similar. One of the key differences is in the world of the passenger carrying operations, the PSU will not do strategic conflict resolution. The PSU will make, with a demand capacity balance, they will limit the number of aircraft entering the airspace to the level that the tactical processes can safely deal with, and then the separation will be tactically achieved by, well, by means which are, still under discussion, actually, because, one of the interesting things about this work on, urban air mobility in the U. S. A. is this idea that there will be these community based rules, and these community based rules are still being worked out. We in Europe are being a bit more top down, we are saying it's going to be the similar approach which was, is being used for small UAS. So we're, at the moment, we're talking about strategic followed by tactical. It may be in future that we might drop the strategic. I don't know. It depends. That's the subject at the moment for research. But, the model that we're working on today would be a strategic with a certain residual risk and then tactical to deal with that residual risk.

Jim: 1:20:34

So Andrew, paint a picture of what the next five or so years looks like in relation to U-Space drone operations in Europe and potentially beyond.

Andrew: 1:20:44

I think, I would really seriously hope that within five years we see, operations of a commercial nature in Europe. I would expect to see a revision of the acceptable means of compliance of the U-Space Regulation. I would expect to see the American Part 108 has been published and is in use. I hope, but I'm not expecting a revision of the U-Space Regulation itself. Regulations are hard things to change. I would like to see, commercial operations. I would expect that they would probably be focused on niche areas where there's an obvious business case. And one of the most interesting areas that we're seeing in Europe is harbours and port areas. A lot of processes in harbours and port areas require somebody to get into a boat, or if you were going to do an aerial observation, It's very expensive with a helicopter. So there are many things that people want to do that they cannot. And so I would imagine, as we're already seeing in Rotterdam, Antwerp, and in, Hamburg Harbor, we will see a growth of use of, drones and, implementation of, U-Space or U-Space like currently, but hopefully official U-Space in Harbor areas. This is, I believe, commercially viable today. The other area, which, is a pressing area of interest is delivery, but delivery today in Europe is tending towards suburban, so it's not really an urban activity. And I would imagine that, the kind of pilot projects that we see in Ireland, in Finland, in Norway, we're working with a company called Aviant, which is delivering around Trondheim in a rural area. We would see these increase in number and we would see them being enclosed inside U-Space airspaces. And we will see these U-Space airspaces popping up like, like mushrooms on a lawn. We will probably be a, not obvious why they're there, but then you go and look and you find there's, here in this place, there's a niche where commercial operations are viable. I would say that the, my least expectation is that it becomes focused on urban, in each case. I would say that probably suburban or harbor will be my initial one.

Jim: 1:23:07

Andrew you've been doing this for a while. I'm sure you have some good advice for our innovators and entrepreneurs on the line.

Andrew: 1:23:13

I would really say work out what your niche is, work out what you're really specialized at. If you, hear some of the stories which are being told about, what people want to do, I would say, is it really realistic to try and compete with ground transport? And you may recall, an experimental project that was done by Airbus, pre COVID, a thing called VOOM, where they set up a kind of on demand helicopter flight service in San Paolo. San Paolo has the benefit of having a large number of helipads, has a quite a large population of helicopters, many of which are sitting idle most of the day, and it also has terrible traffic problems. So there was a willingness among the public, or some members of the public, to pay quite a lot for a short flight and that made this VOOM thing viable. I would say if you're going to come into this business, you need to understand what is the specific set of circumstances where your business will work. And if you come in with a general approach, we're going to deliver cups of coffee. I would say, like really, how can that be better than the man on the bicycle or whatever the alternative is today, you really have to understand what is your value added and where is the niche where that value added is appreciated. And yeah, I wouldn't today be looking at trying to replace or compete with the metro or even taxis in many cities. I'm sorry if I disappoint my colleagues in the taxi business, but really you have to be pretty good to do better than public transport in most European cities.

Jim: 1:24:53

Andrew this has been a great conversation. We really thank you for joining us. What one message would you like to leave with our audience?

Andrew: 1:24:59

I think the message I would leave would be that it's going to be difficult, but the challenge is worth the cost. Where we want to get to is really going to be an interesting place. And so the hurdles are going to be worth climbing over.

Luka: 1:25:13

Thanks a lot, Andrew.

Andrew: 1:25:14

Okay, thank you, bye bye.