Fujitsu uses quantum-inspired algorithm to tackle space waste

The University of Glasgow has worked with Fujitsu and satellite service and sustainability firm Astroscale on a quantum-inspired project to remove space debris.

The project, carried out as part of the UK Space Agency grant, Advancing research into space surveillance and tracking, was developed over six months. It makes use of Artificial Neural Network (ANN)-based rapid trajectory design algorithms, developed by the University of Glasgow, alongside Fujitsu’s Digital Annealer and Quantum Inspired Optimisation Services to solve some of the main optimisation problems associated with ADR (Active Debris Removal) mission planning design.

There are 2,350 non-working satellites currently in orbit, and more than 28,000 pieces of debris being tracked by Space Surveillance networks.

By carefully deciding which debris is collected and when, Fujitsu said the quantum-inspired system, powered by Digital Annealer, optimises the mission plan to determine the minimum-fuel and minimum-time required to bring inoperable spacecrafts or satellites safely back to the disposal orbit.

According to Fujitsu, finding the optimal route to collect the space debris will save significant time and cost during the mission planning phase, and also as a consequence will improve commercial viability. 

Jacob Geer, head of space surveillance and tracking at the UK Space Agency, said: “Monitoring hazardous space objects is vital for the protection of services we all rely on – from communications devices to satellite navigation. This project is one of the first examples of Quantum-inspired computing working with artificial intelligence to solve the problems space debris causes, but it’s unlikely to be the last.”

The project represents the next step in Astroscale’s End-of-Life Services by Astroscale (ELSA) Programme to remove multiple debris objects with a single servicer satellite.

Ellen Devereux, digital annealer consultant at Fujitsu UK & Ireland, said the project not only makes the process much more cost-effective for those organisations needing to transfer and dispose of debris, but it also shows how AI and quantum-inspired computing can be used for optimisation.

“What we’ve learned over the course of the last six months is that this technology has huge implications for optimisation in space; not only when it comes to cleaning up debris, but also in-orbit servicing and more,” she said. “Now we better understand its potential, we can’t wait to see the technology applied during a future mission.”

Amazon Web Services provided the cloud and AI and machine learning tools and services to support the project. The Amazon Sagemaker toolset was used to develop the ANNs for predicting the costs of orbital transfers.

Fujitsu, who spearheaded the project, is among seven UK companies to be awarded a share of more than £1m from the UK Space Agency to help track debris in space. The UK Space Agency and Ministry of Defence have announced the next step in their joint initiative to enhance the UK’s awareness of events in space.