Vigilant Aerospace CEO, Kraettli L. Epperson, presented “New Models for Integration of Unmanned Aircraft Systems (UAS) Into the National Airspace (NAS)” on the second day of the conference.
The presentation is a condensed version of a new workshop Vigilant Aerospace Systems is preparing that will focus on UAS integration into the US national airspace and to airspace management for emerging droneports and UAS airparks.
You can watch the full presentation and read the transcript below (17 minutes and 51 seconds):
“New Models for Integration of Unmanned Aircraft Systems (UAS) Into the National Airspace (NAS): Airspace Management as a Service”
Hello. I’m Kraettli Epperson, the CEO of Vigilant Aerospace and I’m going to be talking about new models for providing airspace management as a service. Particularly, we’re going to be talking about systems that can help emerging programs, small and medium sized programs, which sometimes doesn’t get as much attention, as much resources.
And we’re going to talk about how, [in] the first half of this presentation, I’m talking about our products, FlightHorizon, our history and background in developing and working very closely with NASA. And, in the second half of my presentation, I’ll talk about how we are taking FlightHorizon COMMANDER, one of our products, and using that as a software-as-a-service and airspace manager as a service to help enable programs quickly and get them the types of systems and risk mitigation and other things that they need to be able to launch.
We’ll start off talking about company a little bit and about the products and then we’ll talk about those program challenges. So, we’re a tech startup. We have experienced technology leadership and we have some venture capital firms involved in backing the company. We have offices in Oklahoma City and in Fargo. We’ve got Ryan Aasheim here, who is in our Fargo office. We provide airspace management and detect-and-avoid systems [that] we’ve licensed from NASA patent prototype software and the algorithms to make that work. It includes the hardware integration. It’s exclusively a software solution, so we’re agnostic as to the data and sensors. We have pre-integrations built for a lot of popular sensors and as new things come out, as I’ll talk about quite a bit, we add those to the tool-set very quickly. [It] provides active situational awareness and I’ll show you a quick video here in a few moments.
And some pictures so you can see what I’m talking about with that. [It] provides active detect-and-avoid – either for a pilot to provide an advisory and a specific command to get out of a situation, and maintain all clear, or to an autopilot. That’s really what NASA designed the system for. That’s where we’re heading and we’ll talk a little bit more about that.
The existing infrastructure is designed to be flexible and use future infrastructure and also meet FAA requirements. So, the basics of the product are that the software sits in the middle. On the left hand side here, you’ve got your ownship. And, you’ll be tracking where that is and you’ll be tracking information for other aircraft. The transponder proceed from a radar, from other sources of online data, including, eventually, hopefully UTM and network sources. And then the software uses that to present you with picture. Present you with important alerts, warnings when necessary, and then self-separation advisories with SSA.
The data flows from your sensors, your radar, your transponders, the network, if that’s available, into your airspace model avoidance model, which is continuously running and monitoring the airspace. And then, that information is digested and sent out as needed to the airspace manager and that user interface, out to a pilot or out to an autopilot.
So just a little bit about the way the software is thinking in the background. It starts out detecting [and] parsing data that it’s got, fusing that together, looking at it and creating a picture of the airspace that includes an index of all of the targets. It’s been tracking those targets, building trajectories over time, predicting where the aircraft are going to be. And then predicting potential collisions and it’s doing all of that many times a second. Usually, the latency of the sensors, and other inputs, is much higher than the latency of the software. The software can do this pretty much with as much data as you can give it with a modest infrastructure.
It iterates through strategies determines if there is a self-separation strategy. [It] determines the most efficient of those are then based on some parameters that you might have about a particular airspace or other types of maneuvers you need to make. And, certainly including the performance of your aircraft, which you set up and save in the system when you first set it up. It will then provide that avoidance strategy for you and then it will continuously monitor on a sub-second basis that the aircraft has not gotten back into a conflict situation or that a new conflict has not emerged. And then it will find a new strategy to get you out of that situation.
So that’s really… NASA spent a lot of time on this. We’ve picked it up and have developed a software and commercialized that fundamental process. So, from our point of view, there is a lot of effort right now on detection that’s really important. Obviously there’s a lot of great systems for navigating aircraft. But, there is this big gap that’s going to have to be filled in the industry in order for us to have truly autonomous systems in the future and that’s the ability to do fully autonomous avoidance.
That’s really where this technology ultimately comes in. Although there are several steps between here and we’re using, so when you do use the user interface, this is the 3D synthetic cockpit version of that interface and we’re constantly advancing this. There are new versions of this just since I made these slides. Your ownship will sit in the middle of that so it keeps you aware of where you’re located.
It will be tracking other aircraft and it will show you those even if you cannot see them otherwise. It’s going to be tracking those aircraft even if they are out beyond visual line-of-sight or if they’re behind a hill. It’s going to be tracking those usually by the transponder and then of course as they get closer they also use radar, particularly if they’re not participating in air traffic control. It will then predict if you have potential conflicts. So, it will figure out where that is. It’ll tell you about it and then it will begin producing those advisories.
So, a little bit of background quickly about this. In 2013 NASA started developing the system started doing some basic testing on it. In 2015… late 2015, we licensed this technology. We started doing beyond visual line-of-sight testing with NASA and I’ll go over that little bit more… in 2016. In 2017, we really launched our commercialization process and we actually were called out to help with some disaster response flights. We were able to pull the system in, have it set up very shortly to be able to help with all that. And then 2018 we have multiple things going on.
We’re involved in new programs and we’re about to launch FlightHorizon 2.0, which we’re very excited about. That’s really the second half of my discussion here. In 2018 and 2019, we’re going to be involved in flights were involved in a couple of teams there and we’re going to be learning a lot from that. We’re very excited about that. The new version of the software is routine, but everything is mature and available for our system. So, this system is designed to fly, has been flown with small, medium and large [UAS]. It’s highly parameterized in the sense that we can change it really on the fly, make the settings that we need within certain boundaries to fly with different types of aircraft.
Here’s a quick video of the system in use. You’ll see here. The white aircraft in the middle is ownship. It’s flying through a scenario in which it’s got some conflicts that have been detected. You can see that green one, that’s the computer thinking about and providing a specific resolution advisory – a specific advisory there to get out of that situation and then releasing the autopilot back to the flight path it was originally on.
That green line also provides an audible warnings. It provides visual warnings. It provides text and all that can also be sent out to other systems including, of course, the autopilot. So, it’s a proven visual line-of-sight performer. This is something that NASA designed initially, has tested it and we’ve published an AIAA paper with NASA about this. We had FAA and FCC observers work with us at NASA Armstrong while we were going through this process. It’s a proven detect-and-avoid performer. It’s been through hundreds of encounters now, in which all different types of scenarios have been tested with real aircraft, both manned and unmanned scenarios and we’ve also co-authored a paper with NASA about that.
We had an opportunity, actually, to go out and sponsor and support a team called Humanitarian-Drones.com. They were providing a disaster response flights to FEMA. Helped to produce the image assessment after Hurricane Harvey in Houston. They were able to get 5 eCOAs under the blanket TFR in the region and go out and do those flights during that emergency. So, that was a very educational, very useful process.
Recently, we have been called back in and the system has been subscribed to by NASA for multiple programs – SonicBAT in 2017, ND-MAX in 2018, and in 2018-2019, NASA is using it for… they actually had to have 3 subscriptions for 3 different air management systems for commercial supersonic program. So it’s for formation flying in some cases within ND-MAX, and in other cases, it’s for extensive logging and data collection from multiple sites.
And then, finally, we’re involved in a couple of the IPP teams. We’re on the North Dakota team, which we’re very excited about. And, that’s developing quickly and we’ve already heard exciting stuff about that this week. And then, we’re also on the Alaska team, so we’re going to get a lot of additional flight experience that way. I’m happy to provide information on the background of the technology if you’re interested in that. Most of this is on our website, so you have access to that.
Alright, so we have 3 products. We’re talking about the GCS, which is for an individual aircraft. It’s really for an individual pilots. There is the COMMANDER version which is for air traffic management, or really for airspace management or multiple unmanned aircraft. NASA uses it for all kinds of things. But, we’re primarily focused on enabling small programs to get out there and mitigate their risks and be able to make their safety case. And then, of course, there’s the PILOT version which goes onboard and is intended to talk directly to an autopilot. That’s where we feel the industry is going and is quickly begin to move. So, I won’t go through all of the features of COMMANDER.
We’re going to talk a little bit COMMANDER now. It does provide that 2D & 3D visualization process. It fuses data from multiple sensors. You can use it offline completely over the air if you’re using ADS-B and either a ground based or even an onboard radar, so all of that can be fused together. But, if you have online access and particularly if you’re in a spot where UTM is being tested or eventually deployed, then we will also very much pull that data into the system. It provides a lot of logging and that’s really useful for early stage programs where it is important to be able to do visual line-of-sight flying. Prove that your system works over and over and just be able to have that, replay those logs instantly document that. That’s all built in. When you start using our system, it does that out of the box.
So, a real basic diagram of how it works. You’ve usually got transponders, rules of the road – that’s number one. We use that, obviously, to avoid larger and make sure you’ve got that in your system. Radar is second. Usually that starts as ground-based radar. And quickly there are emerging onboard radars from larger aircraft. We have access to an onboard radar, so we already have modules to pull that data in and begin to fuse it with the airspace model. And then, visual sensors; some of those are emerging. We’re very excited about those.
They offer a low cost alternative to detect a lot of different types of aircraft. And then there are future sensors. There are acoustic sensors, other types of visual, infrared sensors that are emerging and those are great add-ons for our system. So, that’s really how it’s designed and architected. Then that data goes through and then is pushed out to a display or to an autopilot to translate it where can actually be displayed in other systems. So, we can output that. So if you already have existing systems into what you want to put this information, particularly the resolution advisories, we can go ahead and push that out.
It’s sold a software-as-a-service. It’s a package, [that] includes all of the updates and upgrades that you need over time. And, it includes that integration to the hardware.
So I’m going to talk a little bit now about how this is approach to airspace management where you have a piece of software that is architected correctly, can help – particularly small and medium sized organizations – where there’s a lot of complexity that you have to overcome with some solutions. Having a system like this and help overcome a lot of that quickly.
First of all, obviously, programs come in all shapes and sizes. They usually start out small and that’s something that our system is designed to support, designed to support that by allowing him to deploy, you need to deploy initially. To only talk to the systems that you need to talk to and we have trial versions that you can use for additional testing – [to] do visual line-of-sight testing, to set up your system, test your environment and to be able to do integration. We’re finding a lot of interest in that. That’s where we are starting with a lot of people is be able to do that initial integration – maybe with assets that they already have.
The other thing that we find with programs, with medium sized programs, is that they have their particular goals – all types of goals across the industry. So, you’ve got public sector, private sector, mixed public and private partnerships that are developing programs that are supporting the industry, supporting, getting out there with different types of flights. And, what we don’t want is… we don’t want the airspace management system and the safety management and the safety case to drive what the program could do. So, we really want the system to be able to support the program’s goals and not the program have to essentially adapt to the system can do for you. So our system is very flexible and it’s intended to come in small and you can add what you need as you need it as you grow.
Then, of course, there’s regulation. Becoming compliant in the waivers that you need – being waiver ready across multiple uses and growing uses over time for your program is extremely important. This is something that we’ve spent a lot of time focused on. We’re on several of the industry standards setting committees including the ASTM committee on UAS standards, waiver, detect-and-avoid and detect-and-avoid testing, And, so we’re constantly thinking about this.
A lot of what we’re seeing is a trend towards standards that ask for a fused system that is detecting both cooperative and non-cooperative [aircraft] and that is creating an overall picture that presents you and your systems, your airspace management system, and eventually your autonomous avoidance system with a picture of the airspace that is over-detected. It’s detecting all potential targets, hopefully more than once, and fusing them into the system. So obviously, you may have some non-cooperatives where you’re only detecting them once your radar is hitting them and lots of cooperatives where your radar, your transponder, receiver and other things maybe all fusing those together. The idea is that you’re getting very, very high coverage. So that’s a really important part of making the safety case and that’s really where we’re focused with the whole system that we provide. Moving towards UTM in the future as a way to help with that whole process is important.
The other thing that we find related to regulations and waivers, of course, is that small or mediums sized programs, and even large programs, there’s a lot of moving parts. There are a lot of systems and technologies that have implemented at our goal with this product, FlightHorizon COMMANDER is to do as much of that work ahead of time out of the box so you have the hardware integration.
You’ve got recommended hardware compatibility – things that you can pick up off the shelf, plug in and the software is already going to be prepared for that. So the burden of integration is not on you or your program. The burden of a lot of the risk management and risk mitigation and the argument that you need to make about how you’re flying safely and what that means for you is included in the product. And so we’re trying to do as much of that thinking and anticipate, hopefully years or more ahead, where the regulations are going, where the systems are going – including especially UTM – so that our system is aware of all that.
So, costs and complexity. This is a big topic. It’s a topic that I’m sure everyone here thinks about a lot. You’ve got staffing and staffing requirements. You probably don’t have a full time airspace manager. In a lot of smaller programs we run into that a lot. You’ve got training. You’ve got retraining requirements. You’ve got total cost of ownership, of having a program like this and so we think about that upfront. That’s one of the reasons that we price our system as an annual subscription where you get all of our updates. So as things emerge and new hardware becomes popular, it’s going to be included in your subscription and the interface is relatively simple. The system is doing a lot of its work in the background. And so, the point is that the training can be kept to a minimum and you can have people who can get trained and use this system because it is relatively not complex compared to some other things you might try and fulfill these needs.
And then finally capital investment risks. Setting up these programs costs money. We’re very aware of that. We’ll think about that and you want that capital investment to be very justifiable and very future proof. So, you to have made this investment in such a way that you know you can expand it later and those regulations change and new technologies emerge, you’re right in a position to use the existing system resistance movement that you’ve made and expand it outwards to be able to cover the things that you need covered – both regulations and new sensors, for example.
We’re very focused on modularity. Very focused on scalability, extensibility, something that you may have seen in my prior diagram with these systems side by side, and they will talk to each other. You can put an individual ground control station system that an individual pilot is using and they can communicate back. And, the airspace management COMMANDER system can subscribe to that GCS unit. We can subscribe to an edge unit, a pilot unit that may be flying at the edge of your airspace.
That data can all be pulled back and added into your overall airspace model. That might be an important part of your strategy. Be able to watch around with edges and fly around the edges of your airspace and have all of that centralized and logged in your system. So, all of those things help to make the system re-scalable and very extensible. Architecture really matters and this is one of our watch words really at the office and in our software development process – how your software is architected from the beginning. The thought that’s been put into it is going to have a big impact on your ability to continue to use that system and for that system to be able to respond to your needs in the future. So, that architecture is what we think about a lot.
We are sponsoring the luncheon tomorrow. We’ll be able to show you some additional videos and provide additional information. And, I’ll be happy to talk with you if you have questions about this system. Thank you all very much.
Want to hear more? This presentation is a condensed form of a one hour seminar that we are presenting at the upcoming inaugural meeting of the DronePort Network on Wednesday, September 12 in Broken Arrow, OK. Register now to join us! Read more.