NewSpace, spacecraft architectures, and software-defined satellites – with ReOrbit


Episode 36 of the Space Industry podcast is a discussion with Ignacio Chechile, Chief Technology Officer at satsearch member company ReOrbit on NewSpace and software-defined satellites.


Episode show notes

ReOrbit is a Finland-based manufacturer of reusable space systems and in this conversation we discuss a range of topics relating to innovation in NewSpace. In the podcast we cover:

  • What the NewSpace sector is and how it is characterized
  • What makes spacecraft architectures different in today’s missions
  • The importance of software at all levels of a mission
  • The concept of a software-defined satellite and what opportunities this is bringing to the industry

ReOrbit’s portfolio

Episode transcript

Please note that while we have endeavored to produce a transcript that matches the audio as closely as possible, there may be slight differences in the text below. If you would like anything in this transcript clarified, or have any other questions or comments, please contact us today.

Hywel: Hello everybody. I’m your host Hywel Curtis. And I’d like to welcome you to the space industry by satsearch, where we share stories about the companies taking us into orbit. In this podcast, we delve into the opinions and expertise of the people behind the commercial space organizations of today who could become the household names of tomorrow.

Before we get started with the episode, remember, you can find out more information about the suppliers, products, and innovations that are mentioned in this discussion on the global marketplace for space at

Hello, and welcome to today’s episode. Today I’m joined by Ignacio Chechile from ReOrbit Space.

ReOrbit is a Finnish manufacturer of reusable space systems. And today we’re going to talk a little bit about the meaning of NewSpace and how different spacecrafts architectures and the concept of software defined satellites are leading to new opportunities and creating a new discussions and new avenues of progress in the industry.

So Ignacio great to have you here today. Thank you very much for being with us.

Ignacio: Hey yes. Thank you for inviting me. Great to be here.

Hywel: Great. Now your work at ReOrbit I think it’s a really interesting area of the industry. When I say the industry, I think we usually mean NewSpace and a NewSpace is often talked about as a movement to where the cost of access to space is being reduced through some form of innovation, often funded by the private sector.

For you personally, and, speaking on behalf of ReOrbit what does NewSpace mean? Take it into account what you’ve seen happening in the industry recently.

Ignacio: In my opinion, NewSpace is it’s a very broad concept. Of course. There is like a lot of different people doing many different things in a NewSpace, what it’s called NewSpace and what is NewSpace to me, remember, I’m seeing this from the satellite manufacturer perspective.

I’m narrowing it down, through that perspective. NewSpace. In a way, something that I discovered around, like the 2010s, when I realized that it was possible to do space without the NASA logo, like stuck to the missions, without the space agencies, logos, like all over the place, meaning that private people could do space without the big guns paying attention or sponsoring.

And that was to me, the first kind of big changer, the big, like game changer when I came into then. It was clear that it was possible to do space without having like hundreds of millions of dollars in budget, that was also a game changer. Now clearly the big enabler for this to happen was launch costs coming down.

And that was possible by means of the rideshares. The rideshares becoming available, to be possible to flight tourists, to be possible, not to own a full rocket for you, but just, ride a rocket with others that brought the cost down like dramatically that open the door for like many small actors coming and, doing cool stuff and with small spacecraft. And that was the game changer to me.

And that’s what NewSpace means to me in the beginning. That’s how the door opened.

Now NewSpace then became developed in time and it became a bit of constellation, earth observation, change the texture, many technologies that are possible because of the ride share. So the launch costs, like decreasing and I have to say, we need to thank that traditional space for that because the traditional space was the one creating the ride share for the NewSpace to know, be able to fly tourists. So in a way it’s a bit ironic that NewSpace exists in my opinion, because of classic space, we’re making it possible by means of ride share.

But its still the case, because, I mean companies and, startups cannot pay for a full rocket. So still the ride shares are like very much the enabler, so that’s NewSpace, if I see it from my perspective, that’s NewSpace to me.

Hywel: Great. Yeah. That there is an important thing to note that’s NewSpace’s is in debt to the traditional space domain.

Ignacio: We should not forget about that, right? Because somehow NewSpace can have this. Sometimes, there’s a bit of a, or cocky attitude and maybe it’s a little bit of it is thanks to, to classic space.

And that’s fine. We don’t have to, look back a lot to that, but we need to remember.

Hywel: When we are looking back at the traditional space domain compared to things in the NewSpace sector, Specifically in the area of spacecraft architectures, what do you see are the biggest areas of divergence between the two areas? Why do they exist?

Ignacio: That is a huge difference in classic space, spacecraft architectures are extremely what it’s called, like federated. Meaning, one computer that’s one thing. There’s a one to one mapping between the functional architecture and the physical architecture. And it makes sense. It also related to how the company is, or that the space agencies are formulated back in the day, right? Like full teams of people doing one thing. Now I do the control power systems, comms. You name it right.

NewSpace. On the other hand, it came to break a little bit that mapping between the function architecture and the physical architecture. What I’m trying to say here is that one computer that’s not exactly mapped to one function anymore, or that, that is the trend, let’s say. You can have, for example, your attitude control algorithms running in one computer, alone and exclusively, or you can have it, along with many other functions, on, on some, computer as well, and running along with the power system and the thermal software control.

And that to me is it’s a big change. Now classic space, again, these are those of over engineering over there. That there’s good reason for that. Over-engineering come from, high stakes, serious missions that if lost, will create a lot of problems for many people, now that over-engineering is costly and of course NewSpace cannot afford it.

And in a way, NewSpace also came to cut the signal from the noise, let’s say in a way that there was a lot of over-engineering that because of not being possible to be afforded, then NewSpace discarded. And of course it came with some risks, but those risks were taken, I would say, in a bland way, not just like blindly and it gave way to leaner architectures where, as I said again, Computers can be doing more than one function, which also reduces math and reduces complexity of operation, and many other reasons.

Another difference is that, for example, the way you operate and your NewSpace satellite, compared to how you operate the traditional satellite, it sorts of very different. In, in classic space, every bit that you send to the satellite means something. There’s a zero overhead policy. It makes sense because the limited budgets and because of this and the bandwidth that there’s a lot of, reasons behind that. But NewSpace it’s more like treating satellites as, computers connected over a network like a laptop, it’s like a server over a network or across a network. And that brings a lot of facilities. It also brings some problems and some challenges because it becomes a sys admin problem.

It’s like a system administration issue more than our satellite operations issue, but it really streamlines the operation of a remote system like a satellite. As I said, maybe to summarize this question on one hand, traditional space, highly federated, highly specialized equipment. On the other hand, NewSpace architectures where the roles of computers can be, it can have like different functions and it can also change with time because you can reconfigure the computers after you launch it and say, okay, now you are not doing anymore.

You are now you are taking care of the, the power system software or something like that. So that is the, that’s how I see these two things, right? Of course the division, sometimes it’s blur. But if you ask me, what is the colours of these two sides, classic versus NewSpace in terms of architecture. This is how I see it.

Hywel: Brilliant. Yeah. So it’s about an increase in versatility. There are drivers for that, as you say, there, the technological ability to actually develop it, but also on the demand side, satellite data requirements are changing and the market will change. And if satellite systems are able to adapt to that faster, then the opportunities will increase.

Great. That’s really interesting. Thank you now. And towards the end of the edge of the innovation envelope in this area is the work that you are doing at ReOrbit on software defined satellites. What does the concept of a software defined satellite mean in NewSpace?

Ignacio: I hope you have some time because I get a bit passionate!

Hywel: Of course.

Ignacio: There is one ugly truth that I have seen in many different missions and in companies, that I’ve been working with them and there is an ugly truth that finally someone has to say open and loud. In satellites and in space, software rules. I know a lot of great mechanical engineers.

I know a lot of great, thermal engineers and a lot of, electronic engineers. Yeah. They are great people, but in a space mission projects, the software guys are the guys with and why, because it’s related to the previous question as well. When you have an architecture that you can reconfigure and you can, define the different roles as the mission evolves and all that, it is all enabled by software. There, there’s no way you can change bolt from one place to another place easily in space, but you can change software. Software is your asset software is what can, allow you to say, no, you’re not doing this anymore. You’re doing that. You can change the functional, you can change the roles.

You can change how data flows in, what direction, how it’s stored. Software is the real assets and companies ignoring this, they end up creating architectures like that as well. Coming back to the previous question, I repeat is that they create architectures that are ignoring this fact, that software is a real asset and they should on their own feet because then they launched satellites that are very rigidly, like configured to do one thing.

Since I joined ReOrbit with a year and a half ago, we’ve been brainstorming a lot about the software defined concept because there is a lot of, it’s a big bus for us. People just use it like very loose. Oh yes. software defined this, software defined that, what it means for us software defined is that when you have a remote system, you need to operate a remote system.

It can be a satellite but it could also be a, a nuclear plant or it could be a, a power plant or similar, right? When you have a remote system that needs to be operated, you are interacting with the software. That’s what you talk to. That’s what, talks back is the software.

Then when your architecture is a little bit decouple, then you always talk to one actor in this architecture, this remote architecture, you talk to the, the onboard computer engineer. And then the computer goes like through different immuno pains and issues to, go and talk to some other, slave computers to send commands back and forth is a bit of a mess. So what we’re working on here is that, okay. A satellite, it’s a collection of computers and we are not going to change that.

You will still have like several computers. Now we are saying that these computers, they share a bit of a common, it’s like a common, memory space, even though the memory is not really physical because they don’t have, they don’t share the same the memory chips, but they share a satellite. They are all flying on the same object. And as such, they need to share data and they need to be aware of data from other computers in the satellite. So we created a bit of virtual memory space and we say, okay, now all the computers are aware from each other.

And they all the computers, if they need from, data from some other computer, they just ask for it because they know that, Hey, this number that I need is owned by my, my fellow computer connected through my CAN bus or my space wire bus. So we would configure the satellite in that way that we don’t really mind that we don’t really think about the interfaces or we don’t really pay a lot of attention to that space wire drive.

Of course, we have to have space wire drivers and CAN drivers and, you are drivers because you have to, but what we care about is the data, data onboard, how the data flows, but for us, software basically handles data. If you don’t really know what your data is, then there’s nothing you can do onboard of a satellite.

And we extend that concept to the ground because from the ground perspective, the operator also needs to get data and send data. Command is data, telemetry from a satellite is data, and we make that data flow as seamless as possible because at the end of the day, when an operator is doing by sending a command, they want to modify a position in memory of some computer on the satellite, whatever that is.

And we have created set of abstraction layers that we make sure that the operator doesn’t really need to care too much about where exactly the data is sitting on. It’s more like I need this. Give me the data and that at the end of the day, just to put a little bit of practical coating on this, it is a software abstraction layers.

So we grade those abstractions that from the ground, the operator doesn’t need to care and why we do this because also we’re looking a little bit ahead. Because eventually we want at ReOrbit we are trying to make satellites also, fly together and network together in orbit. And we also want to extend this concept to the satellite constellation.

When you have data on similar satellites, imagine that you have a formation of 10 satellite flying over your head. You can not operate each one of those satellites individually. So you need to create a bit of a sense of a flock. And that’s why we are just looking forward and saying, these abstraction layers would eventually in the future, allow an operator formation of satellites passing over their heads. And that’s what software defined means for us. We want to disect the problem of operating a remote system without the space hallowed, something that’s had been launched for 64 years already. We don’t buy that hollow anymore. I It’s like satellites are satellites. Satellites have been like that for like decades.

Let’s just move on and think about, how to operate them in a proper way. Sorry for the long answer. I said, I get passionate.

Hywel: No, that’s really interested. Really clear. Yeah. Yeah. Looking forward to those applications that you talked about, for example, and then as you say, in fact, the launch costs have come down, which you’ve mentioned, and we’ve seen in other areas, the miniaturization of subsystems, it means that you’re able to launch more cheaply satellites with a greater number of sensors and greater capacity to do all sorts of different applications, but determining how you manage those applications and determining what a task a satellite to do is a matter for the software and the operator to do so. That’s really interesting.

And in this area, however, there are a lot of buzzwords that are thrown around. Including and, to greater or lesser levels of applicability, but there are AI on satellites as an example, distributed architectures, in-orbit networking, and as you’ve mentioned, software defined technologies, and you’ve alluded some of these, like the use of thinking of satellites as servers in space and that sort of thing, how much you are such technologies in your view, and particularly in their ability to provide, services with good reliability to end-users on the ground?

Ignacio: Yes. buzzwords are a big problem in this industry. We were talking about classic space and NewSpace in the beginning. And for some reason, unfortunately, the NewSpace industry, perhaps coming from the startup scene adopted this policy of buzzword s all over the place.

And that’s a big change from classic space. The missions do the talking. And in NewSpace, there’s a lot of marketing going around, and that is a bit of a problem because some of these technologies that you mentioned, for example, AI or software defined architectures or, distributed architectures, people just talk about them in a very empty manner.

And then they start to lose the meaning because they it will have to take them as just yet another far and other, like buzzword that people are just like spouting. And that’s a problem. Now, if we remove that, don’t know how, but I guess that in the future, less power to marketing teams, more power, but we know products that’s maybe one, one way to go.

But anyway, maybe don’t get up all those technologies and how much mature they are. If you think them individually, AI, the distributed architectures and software defined architectures, they are mature on their own on the ground, let’s say so on the ground AI, it’s of course it’s maturing as we speak, at the stage of the, but still it’s already, you can really see the value already of AI. When it’s done in the proper way, right? Not just a list of nested if statements, but more like proper AI, it’s already paying off. Distributed architectures, I don’t really need to give a lot of examples on the ground.

Of course, there are many examples of that distributed architectures or the centralized architectures, the blockchain and all these things that are, are becoming like very topical are already rule for that, there’s a possibility of having distributed systems without a single central entity and still being able to add value. And networking I’m not going to talk about, how networks work on the ground because, we, we are now talking, talking here thanks to network. So I’m not going to explain, that networking on the ground is mature now how to bring that to orbit. I think that this is something that is you need to gain confidence in this technologies in orbit by means of taking the proper steps.

Meaning that I don’t think you need to go from zero to everything. I don’t think that you can go from no AI in space, no distributed architectures and no in-orbit networking all the way to having all those features, working upfront. You need to take them slow, and slow I’m not saying in decades, but more like. You need to prepare, a set of pathfinder missions where you say, in this pathfinder mission, I will showcase the in-orbit networking feature. In the next pathfinding mission that can be just few months away and with very small satellites and cheap satellites, you can say, I’m going to try, distributed architectures.

And here at ReOrbit, we’ve been discussing this, you know how, because we are pursuing all these things. We are pursuing all this. How do we make the industry aware that all this is possible. Again for an industry that is, historically a little bit conservative because of the heritage from classic space.

So we say, we would say that, we need to walk the talk. There’s no way that you can promise in space that you would say buy from me because this would work because I say it, you need to show that it works right. And that’s why we are envisioning a set of Pathfinder missions.

In this particular mission, we’ll try in-orbit networking. And that means launching two small satellites with the capabilities of talking to each other and then exchanging information in specific protocols, being able to change those protocols and show that, Hey, you can have two satellites in orbit if the same way as if they would be like, like network node. And the same for the rest. For AI, we are working internally here on self diagnostics making sure that satellites can have running in our statistical models onboard or assess the status of the power systems, of the battery, but how the battery will deplete, according to some scheduling where we’re doing all that, but we are going to showcase those in Pathfinder missions. And from those Pathfinder missions. We are going to show the market and the industry, which again is a bit conservative to say, Hey, this was not just empty dock.

They would do. It’s not just small more smoke and mirrors as you will find all over the place. This actually works. And we will also equip our product line with these features as we show that the advanced features work. And I guess that’s something that we cannot fully overcome, walking the dog requires going into taking proper steps and that, the capability of launching cheap satellites, small satellites for cheap launches or on cheap launches is it’s a good path, to increase, TRL or, the technology readiness.

Hywel: Absolutely. And yeah. You mentioned that flight heritage in many ways is just binary. It needs to be done.

Ignacio: Yeah. And also you mentioned you at some point, you said reliability. I think you said the how you show this, in a reliable way. And when it’s of course attached to what I said that way you need to go slow and show it. But at the same time, reliability in my opinion is an architectural problem. It’s we assume here internally that things will act up in space because the space is hard. Now your architecture needs to be ready to overcome that things that will definitely happen because they will happen.

And I’ve been part of too many missions already to realize that it’s going to happen. So you need to make sure that your architecture is ready to do the switchover, do the configurations to make sure that you can continue providing service after the shady things would happen because space is space and we cannot change that.

Hywel: Yeah, so you’re designed and for the environment accepted the limitations that it brings. So that’s great. That’s covered a lot of the questions and things that, that that I have for you today, I guess just finally, I always ask a form of this question to all of our guests, looking into the crystal ball of the future, I wondered where you thought you saw different forms of activity in the NewSpace sector heading in the next, five to seven sort of years. What are you most excited about as well at ReOrbit?

Ignacio: That’s a good question. I cannot really do a lot of, crystal balling because also people will, replay this podcast, in a five year from now.

And they will laugh at me because all my, my forecast for more wrong, wouldn’t be the first time. So the good way of putting it this, what is it that I’m looking for in this industry? I’m looking forward to the trend of commoditization of space equipment to continue, meaning that, that sensors, actuators, computers for satellites, they should stop being special things and AI and very specialized equipment with very long lead times and, and handcrafted natures.

I would love to see that in the future, you can go to a place like satsearch and say, I want to buy. Then start truckers for my missions and I will have them here next week.

I know that it’s, if there’s a long way for that, I think we are on the right path for this. So commoditization of equipment, I hope it will continue evolving in the way that it is evolving today. I would also love the space, as I said, a few questions ago that the space hallow that there’s this aura that is covering space things and science fiction is to me, something that we’re not working on science fiction here, we’re working on, satellite that are just a bunch of computers, connected like cars are, like a Tesla is. And that’s why I want space.

Continue the greasing that, that space hallow that somehow, we move forward overcoming the fact that yes, there are rockets and southerners going to space and that’s that, I mean that we’ve been doing this for decades, so let’s just move on, the process of hiring people and, like scaling companies a little bit easier because you need to start bringing in people from space.

You can just bring people from, automotive. And that’s that. So that’s a trend that I would like to continue seeing. I also would like, and this is perhaps related to my work here and what we are doing at real, with that, I would like software marketplaces to be flight software marketplaces.

If we’re flying a satellite and you need a library for encrypting your data, or you need a library for, I don’t know, a specific protocol, or do you want to have a driver for a specific interface that is a little bit uncommon or something I would love. This trend of having a place that you can go and select what you need.

And then you will be able to, not very much you to installing the software in your satellite, but at least with some steps, already sorted out. So that’s a trend that I would like to see and maybe last but not least, we commented a little bit on the, increasing, TRL before I, I was talking about this in the previous chat.

I think there is a bit of problems in the industry, and on new actors how to increase the TRL in in a more I would say reliable way. A clearer path to increase your TRL because so many companies have great ideas, but they face the famous heritage problem. Now you don’t have heritage and then you’re just kicked out of the discussion because you are not in space. That to me is a bit silly because it’s leaving a lot of great companies with great propositions and offers from, during the conversation because they haven’t been in space. So they need to have a bit of a better way of, technology maturation. And this is something that I would like to see in the next, three, four years that there will be a solid way of new actors, gaining heritage and then, maturing from there. So those are the things that I would have by us, by my work here, but that’s my opinion.

Hywel: Excellent. Thank you very much for, yeah. That’s some really interesting areas that we shared hopefully, but, maybe see in a, in the future and I’ll be really interested in for yeah, the audience to think about. And yeah. Thank you very much for the discussion today, Ignacio.

I think our listeners will have learned a lot about why the different concepts of NewSpace hardware and computing exists today, and the opportunities that software defined satellite concepts could bring. So thank you for sharing.

Ignacio: Thank you for inviting me.

Hywel: Absolutely. And to all our listeners out there to find out more about ReOrbit’s work and portfolio you could view the company supplier hub on satsearch.

On the platform, you can make requests for information, technical documents and other procurement requirements like lead times or quotes or whatever else you might need for the development of your missions or services in the NewSpace sector and beyond, of course. So thank you very much for spending time with us today on the space industry podcast.

Thank you for listening to this episode of the Space Industry by satsearch. I hope you enjoyed today’s story about one of the companies taking us into orbit. We’ll be back soon with more in-depth behind the scenes insights from private space businesses. In the meantime, you can go to for more information on the space industry today, or find us on social media if you have any questions or comments. Stay up to date. Please subscribe to our weekly newsletter and you can also get each podcast on demand on iTunes, Spotify, the Google play store, or whichever podcast service you typically use.

software-defined satellite
supply chain

related articles

Blog home

Microsatellite and CubeSat platforms on the global market

CubeSat thrusters and small satellite propulsion systems

Ground station service providers: an overview of telemetry and telecommand communication services and networks for small satellites