Astroscale isn't really a space junk removal company. Inside its plan to build roadside assistance for orbit

I've been wanting to understand space a lot better.

Weather forecasts, the blue dot on your map app, a video call with someone on the other side of the planet — a surprising amount of daily life runs on satellites circling the Earth. Yet I realized I had almost no idea what is actually happening up there.

The first thing that hits you when you start digging is the scale. Japan's government, in its Basic Plan on Space Policy, wants the country's space industry to double, from ¥4 trillion in 2020 to ¥8 trillion (roughly $53 billion at ¥150 to the dollar) by the early 2030s. It has also launched a Space Strategy Fund meant to put around ¥1 trillion to work over ten years. Space has stopped being a dreamy story about rockets. It has become a huge industry that governments are betting real money on.

More satellites also means more space junk: dead satellites, spent rocket parts and the fragments they shed. The European Space Agency's Space Environment Report 2026 estimates that about 1.5 million pieces between 1 and 10 cm are circling the planet.

The deeper I went, the more one question nagged at me: what happens after? On the road, if your car breaks down you call for roadside assistance, and if you run out of gas you go to a gas station. In orbit, none of that existed.

The company trying to build that missing aftercare for space is Astroscale, headquartered in Kinshicho, Tokyo, and listed on the Tokyo Stock Exchange's Growth market since 2024. And this time, we got to talk to them directly.

When Astroscale makes the news, it's usually for cleaning up space junk. Yet the first thing our interview set straight was exactly that label: "the space junk company."

The short version: Astroscale's real business is roadside assistance for orbit — getting up close to satellites to inspect them, refuel them, extend their lives and, at the end, remove them. Clearing debris is one item on that menu. What ties it all together is a single, very hard skill: approaching something in space that won't cooperate.

WHO IS

Satoshi Ito

Public Relations, Astroscale Japan Inc.

“Debris removal is where we started. It isn't our main business.”── Satoshi Ito, in this interview

1. Early careerPR agencySeveral years handling public relations for corporate clients
2. 2021Joins AstroscaleMoves into space PR with no background in the industry
3. NOWPR, Astroscale JapanPart of a five-person team running announcements and trade shows

Ito started his career at a PR agency. In 2021 he joined Astroscale, just as the company was setting out to make on-orbit servicing a normal part of how satellites are run — with zero background in the space industry. Today he is part of the PR team at the Japanese subsidiary, handling mission announcements and trade shows.

"The space industry" covers an enormous amount of ground. Saying "space industry" is a bit like saying "Earth industry." There are companies heading to the Moon, companies building satellites, companies using satellite data, rocket companies.

We work in a field called on-orbit servicing. It's not a well-known term, so the easiest way to put it is roadside assistance for space — think of Japan's JAF, or AAA in the US, but in orbit.

Space may be infinite, but the altitudes where satellites are useful are not. More and more satellites are being launched into those altitudes, debris keeps piling up, and orbits are getting congested. The result is more near misses and more collisions.

For satellite operators, near misses and collisions are nothing but risk and cost. Right now, space is a place where the risk is going up and the return is going down. Our job is to protect those orbits so future generations can keep using them.

Exactly. If you get a flat tire, roadside assistance comes to you. If you run out of fuel, you go to a gas station or get towed. When a car reaches the end of its life, it gets scrapped. Cars and home appliances already have that whole aftermarket built around them.

Space never had one. You built a satellite, launched it, and when its mission was over you simply left it there. There was no technology to refuel or repair it, so there wasn't much choice.

It's like buying a luxury car, abandoning it on the roadside when the tank runs dry, and then buying another luxury car. That's what's been happening for decades.

At the start, the focus was almost entirely on removing debris that was already up there. But as we went along, we realized that wouldn't be enough on its own. The business gradually grew out from there.

Within on-orbit servicing we have three service lines: inspection, servicing and removal.

Removal is the easiest to grasp, but it isn't the main one. To be clear, debris removal is where we started, but it is not our main business. We'd like people to see all three as equally important.

Servicing covers things like refueling satellites, repairing them and extending their lives. Inspection means getting close to an object and understanding what state it's in. Something a few hundred kilometers up — how badly has it degraded? Is it tumbling? You can't tell unless you go and look. And if you don't know an object's condition, you can't service it at all. That's why going to look is the foundation underneath everything else.

For one thing, you wouldn't need to launch a satellite with a full tank. You could fill it halfway and save on launch costs because the satellite is lighter.

You might also design a satellite for 25 years instead of 15. Once you can remove satellites or extend their lives in orbit, the way satellites are operated starts to change.

ADRAS-J was a slightly unusual mission. It was a demonstration, and it was both an inspection service and a forerunner of debris removal.

What ADRAS-J did was approach and observe. And the debris it observed is the very object we plan to remove a few years later. So you could call it a scouting run for the removal mission.

Satellites basically just go around at a set speed. In car terms, they never touch the gas or the brakes. They make small adjustments to stay pointed the right way, but they don't suddenly speed up or slow down — so normally, a satellite simply can't approach another object.

Removal and servicing, though, are impossible without getting close. ADRAS-J was the first mission in the world to actually pull that off with a real, large piece of debris. There was no precedent to learn from, so we had to build it from scratch.

Speed actually isn't the issue. Debris and satellites all travel at 7 to 8 km per second, and once ADRAS-J is in orbit it moves at the same speed. The real problem is that the debris doesn't control its attitude and doesn't tell you where it is. It's a target that won't cooperate — that's the big factor.

Honestly, all of it was hard. But if I had to pick one thing: building a satellite is a constant tug-of-war over limited size and mass. If you want more fuel, the body has to get bigger. If you need more power, the solar arrays have to get bigger. The power team, the propulsion team — everyone has their requirements, and the hard part is pulling them all into one clean design.

To put it more simply: because the target won't cooperate, the spacecraft has to think for itself and make its own approach. For that you need eyes, a brain and legs.

The eyes are the sensors and cameras that pick up the target. The brain uses what the eyes see to decide how to get closer — we ran more simulations than I could count and built our own algorithms to work out how to approach precisely and safely. And the legs are the propulsion system that provides the push to get there. Nothing like it existed, so there was no right answer to copy.

That's a great question. There may be some of that, but I wouldn't call it bottom-up.

ADRAS-J was never a one-off mission. It's one piece of a technology roadmap that runs over years — even a decade. Getting close to an object and understanding its condition in detail is common to debris removal and life extension alike. ADRAS-J wasn't meant to stand on its own; it was built to be reused in the missions that follow, so the teams worked together on it rather than optimizing just for this one mission.

The contract came in 2020, and the development time was much shorter than for a typical large satellite. Building something brand new, fast, was genuinely hard.

It really does. The up-front investment is huge, and revenue usually only comes in after a mission. The revenue lands a long way down the road, and that is a challenge.

We've planned our fundraising carefully since the company was founded. But the fact is, we are still running at a loss. We've gotten here with a mix of fundraising and government grants.

Space is unusual: first of all, it has no borders. Territory, territorial waters, airspace — none of that applies. On top of that, there are no lines between sectors like commercial, civil government and national security. Because it belongs to no one, every player operating there is a potential customer. You can't protect the space environment by working with just one sector, so defense naturally comes in.

The same thinking is behind our offices around the world. No single company or single country can solve the debris problem. The usual path for a Japanese company is to grow at home first and then expand abroad, but that would be too slow. We now have bases in five countries, and each team builds its own spacecraft and wins its own contracts.

Structurally, I don't think we're that unusual. If there's one thing, it's that our official working language is English. We use plenty of Japanese too, of course.

Part of the reason is to make it easier for people from overseas to join — our group COO, the No. 2 in the company, is from overseas as well. Above all, we just don't have enough engineers. At each of our bases, roughly 70% of the staff are engineers, and it's still not enough, because every base is working on several contracts at once. We're always hiring.

We break our job descriptions down into very specific roles. If we only hired people with space-industry careers, there simply wouldn't be enough of them. The pool is far too small.

So we split the work up. Take data analysis: you don't need a space background for that. Once you break things down far enough, the industry stops mattering. We have people from the auto industry, which is a natural fit for space, people from industrial plants, people from electronics makers.

Space used to be a fairly closed industry. That's exactly why it's refreshing when people from cars and other fields bring different methods and ideas. Because they cross industry lines, there's a lot we can learn from them.

On-orbit servicing is a new field, and building the technology isn't enough. Without customers, it can't be a business, so we have business development, and teams like ours in PR. Even with the technology and the customers, you can't operate without rules for space — so we have a policy team that contributes to how those rules get made. A lot of the teams are small and highly specialized.

Nothing dramatic. I spent a few years at a PR agency straight out of university, and when I started looking for an in-house role, I happened to come across this company's opening.

It was 2021, right when the company was setting out to make on-orbit servicing the norm. I thought, "Now that's a PR job worth doing," and applied. I knew absolutely nothing about space.

Above all, how much explaining it takes. You start with "what's an orbit?" Most people can't picture a satellite. Say "on-orbit servicing" and you can see the question marks appear.

To make it easier to explain, around 2021 we started leading with the debris problem — space junk. But that image stuck so well that we became "the debris removal company."

In reality, debris removal isn't all we do, and it isn't our main business. So now, when we put things out, we emphasize on-orbit servicing, or say "we're a roadside assistance company for space" to make it a little easier to picture. Space touches everyday life, but indirectly — it's not something people think about. That may be the hardest part.

Decades ago, the goal was getting to space. Now, using satellites in space is part of everyday life. Next, I think an entire economy will form in space, and on-orbit servicing is what will hold it up. It's something people won't notice day to day, but it will be one of the things quietly holding everything up.

The debris problem is urgent. Nobody knows when chain-reaction collisions between pieces of debris could start, or when an orbit could become unusable. That's why we're moving fast. By the 2030s, we want on-orbit servicing to be something people use as a matter of course. In Japan alone we have three missions in development, with two launches planned for 2027 and Japan's fiscal year 2027. Keep an eye on those.

Ito called near misses and collisions "nothing but risk and cost." Put numbers on that cost, and it's substantial.

SpaceX, which runs the Starlink satellite internet constellation, reports its collision-avoidance maneuvers to the US Federal Communications Commission every six months. According to a Space.com report, SpaceX made 207,152 of them from December 2025 to May 2026 — more than 40 per satellite per year, meaning each satellite is dodging something almost every week.

Dodging isn't free. An ESA article on automating collision avoidance notes that avoidance maneuvers "often use up scarce fuel, decreasing the lifetime of the mission". Stop dodging, though, and a calculation by Thiele and colleagues puts the window before a catastrophic collision at just 2.5 days. Today's orbits hold together only because everyone keeps burning their own lifespan to stay out of each other's way.

Seen this way, it's clear why Ito pushes back on "the debris removal company." Clearing away junk and adding fuel to extend a satellite's life look like separate businesses, but they are two sides of the same problem. Clean up, and satellites dodge less. Refuel, and they win back the life they spent dodging. Neither one alone fixes a crowded orbit.

When Ito talked about eyes, a brain and legs, I immediately thought of self-driving cars. Japan's transport ministry, in an interview in its web magazine, describes autonomous driving as a system that takes over "perception" of its surroundings, "prediction" of what happens next, "decision" on what to do, and "operation" of the steering and brakes. Eyes are perception; the brain is prediction and decision; legs are operation. Rendezvous in space breaks down the same way a self-driving car does.

That also explains why Astroscale slices its job descriptions so finely. "Space work" only matches people with space experience. "Estimating a target's position and orientation from images" matches people from driver-assistance systems, robotics and factory inspection. CEO Nobu Okada made a similar point in a 2022 interview with the Japanese space-business site Sorabatake: there are plenty of people "who can strip noise and bias out of data" outside the space world.

It reminded me of our interview with an image-analysis engineer using drones and AI to check crops. The eyes that count rice ears in a field and the eyes that measure a rocket stage's orientation in orbit are close relatives. Data analysis, image processing, control, simulation — your specialty could become the eyes, the brain or the legs of something in space. That's my takeaway from this interview.

So who pays while revenue is still a long way off? According to Astroscale's results presentation, the company's order backlog stood at about ¥36.3 billion (around $242 million) as of July 2026: roughly ¥27.4 billion from civil government, ¥8.5 billion from defense and ¥0.4 billion from commercial customers. Nearly 99% of it is government work.

It would be a shame to read that only as "there's no commercial demand yet." New transport networks have usually been born with the government as the first customer. In 1925, the US began contracting airmail out to private airlines. According to the Smithsonian National Postal Museum, early carriers earned up to 95% of their revenue from mail and could not have survived a year on passenger fares alone. In space, NASA paid SpaceX $396 million through its Commercial Orbital Transportation Services program — and the rocket that grew out of it has since carried more than 10,000 Starlink satellites into orbit.

I think on-orbit servicing is now in its airmail era. Governments buy first, and the technology and track record build up. What I'll be watching for is the moment commercial satellite operators start paying for inspection and refueling with their own money. As of August 2026, one commercial deal worth around ¥7 billion (about $47 million) was at the bidding and contract-negotiation stage, Astroscale told investors. The ones feeling the cost of a crowded orbit most sharply should be the commercial satellites dodging debris every week.

Launch it, use it up, leave it behind. For decades, that's how space worked. What Astroscale is trying to do is add a missing step: maintenance.

Get close, take a look, refuel, clean up. ADRAS-J's 15-meter approach was the first step for all of it. In 2027, the inspection spacecraft ISSA-J1 and ADRAS-J2, which will grab a rocket stage and bring it down, are set to launch. Roadside assistance for orbit is about to move from scouting to real work.