GITAI’s Autonomous Robot Arm Finds Success on ISS

Late last year, Japanese robotics startup GITAI sent their S1 robotic arm up to the International Space Station as part of a commercial airlock extension module to test out some useful space-based autonomy. Everything moves pretty slowly on the ISS, so it wasn’t until last month that NASA astronauts installed the S1 arm and GITAI was able to put the system through its paces—or rather, sit in comfy chairs on Earth and watch the arm do most of its tasks by itself, because that’s the dream, right?

The good news is that everything went well, and the arm did everything GITAI was hoping it would do. So what’s next for commercial autonomous robotics in space? GITAI’s CEO tells us what they’re working on.

In this technology demonstration, the GITAI S1 autonomous space robot was installed inside the ISS Nanoracks Bishop Airlock and succeeded in executing two tasks: assembling structures and panels for In-Space Assembly (ISA), and operating switches & cables for Intra-Vehicular Activity (IVA).

One of the advantages of working in space is that it’s a highly structured environment. Microgravity can be somewhat unpredictable, but you have a very good idea of the characteristics of objects (and even of lighting) because everything that’s up there is excessively well defined. So, stuff like using a two-finger gripper for relatively high precision tasks is totally possible, because the variation that the system has to deal with is low. Of course, things can always go wrong, so GITAI also tested teleop procedures from Houston to make sure that having humans in the loop was also an effective way of completing tasks.

Since full autonomy is vastly more difficult than almost full autonomy, occasional teleop is probably going to be critical for space robots of all kinds. We spoke with GITAI CEO Sho Nakanose to learn more about their approach.

IEEE Spectrum: What do you think is the right amount of autonomy for robots working inside of the ISS?

Sho Nakanose: We believe that a combination of 95% autonomous control and 5% remote judgment and remote operation is the most efficient way to work. In this ISS demonstration, all the work was performed with 99% autonomous control and 1% remote decision making. However, in actual operations on the ISS, irregular tasks will occur that cannot be handled by autonomous control, and we believe that such irregular tasks should be handled by remote control from the ground, so we believe that the final ratio of about 5% remote judgment and remote control will be the most efficient.

GITAI will apply the general-purpose autonomous space robotics technology, know-how, and experience acquired through this tech demo to develop extra-vehicular robotics (EVR) that can execute docking, repair, and maintenance tasks for On-Orbit Servicing (OOS) or conduct various activities for lunar exploration and lunar base construction. -Sho Nakanose

I’m sure you did many tests with the system on the ground before sending it to the ISS. How was operating the robot on the ISS different from the testing you had done on Earth?

The biggest difference between experiments on the ground and on the ISS is the microgravity environment, but it was not that difficult to cope with. However, experiments on the ISS, which is an unknown environment that we have never been to before, are subject to a variety of unexpected situations that were extremely difficult to deal with, for example an unexpected communication breakdown occurred due to a failed thruster firing experiment on the Russian module. However, we were able to solve all the problems because the development team had carefully prepared for the irregularities in advance.

It looked like the robot was performing many tasks using equipment designed for humans. Do you think it would be better to design things like screws and control panels to make them easier for robots to see and operate?

Yes, I think so. Unlike the ISS that was built in the past, it is expected that humans and robots will cooperate to work together in the lunar orbiting space station Gateway and the lunar base that will be built in the future. Therefore, it is necessary to devise and implement an interface that is easy to use for both humans and robots. In 2019, GITAI received an order from JAXA to develop guidelines for an interface that is easy for both humans and robots to use on the ISS and Gateway.

What are you working on next?

We are planning to conduct an on-orbit extra-vehicular demonstration in 2023 and a lunar demonstration in 2025. We are also working on space robot development projects for several customers for which we have already received orders.