��������Vlog

Overcrowding, waste disposal and rubbish have created a huge and growing pollution problem all over planet Earth.

This is becoming a problem off-planet too, with thousands of man-made objects orbiting Earth, including satellites, space installations, space debris and junk.

As more nations and commercial companies develop space programs there is an increasing need to monitor space traffic to try to prevent collisions and damage to trillions of dollars worth of technological investments.

There are more than a thousand operating satellites in orbit and one or two of them are lost in space crashes each year.

A multidisciplinary research team is on to the problem, working in collaboration with industry to develop a space-based surveillance system.

They aim to deploy satellites that use optical sensors to detect objects in space, increasing the capability and utility of space situational awareness.

The team, led by our own Dr Tat-Jun Chin, recently won a global challenge hosted by the European Space Agency.

They used a unique combination of machine learning and 3D vision algorithms to determine the most accurate orientation of an object in space, edging out 50 competitors from some of the world’s most prestigious universities and space technology companies.

"Figuring out the orientation of an object is a long-term study problem in computer vision and AI,” Dr Chin says.

“If you want to program a robotic arm to make coffee, you need to figure the orientation of the object with respect to the robot; we are now applying those techniques in space.”

Our team’s research forms an important component of the growing local space industry.

It could be the foundation for new technologies to remove space debris, or to refurbish and prolong the life of ageing space assets and prevent them adding to space pollution.

It could even facilitate development of space depots; jumping off points for distant space travel.

But all the exciting possibilities begin with traffic management through space situational awareness.

��