Get all your news in one place.
100’s of premium titles.
One app.
Start reading
The Canberra Times
The Canberra Times
National

How can we improve plastic recycling?

In Australia, we need to kick soft plastics to the kerb. Picture Shutterstock

Australians throw away a whopping 3.5 million tonnes of plastic each year but less than a 10th is recycled. When you break that down to soft plastics (plastic bags, cling film) we throw out 70 billion pieces a year - around 3000 pieces per person.

Less than a quarter of Australians are aware that soft plastics are recyclable. The diligent minority who regularly took their old plastic bags to supermarket drop-off boxes have recently been disappointed. The REDcycle debacle has shown that relying on commercial enterprise or government intervention alone is not the solution.

While several councils have implemented single-use plastic bans, we need a whole-of-society system that harnesses advances in AI and robotics to recycle soft plastics that have not - or cannot - be phased out to shift our throw-out economy to a circular one.

We also need to reconsider our local garbage collection. Put simply, we need to kick soft plastics to the kerb.

Not everyone can easily travel to their local supermarket to drop off plastic waste. Allowing soft plastics to be recycled through council collection would reduce this impediment and help increase recycling rates.

But the current issue is soft plastics cannot be placed in kerbside recycling bins because they get caught in conveyor belts, causing mechanical failure.

AI is one solution to this technical challenge. Alongside circular industry companies CurbCycle and iQRenew, our University of Sydney team is developing advanced AI-enabled waste-sorting robots.

In our current pilot program with local councils, registered households collect soft plastic waste in specially designed yellow bags and place these bags directly in recycling bins once full, together with other recyclables.

Once the waste arrives at a waste management facility, our AI-enhanced robots detect and separate the different types of rubbish, separating soft and hard plastics. The system is autonomous and can operate 24/7.

Our approach relies on neural networks - AI systems that constantly learn and are trained on extensive image data sets - to detect and distinguish soft plastics.

After being separated from other waste, these soft plastics become commodities and can be recycled into oils and valuable chemicals, diverting soft plastics from landfill and paving the way for Australia's circular economy.

Putting this idea into practice nationally will need a push from the government, the goodwill of households, and an investment in innovation by waste management companies, but it would far better than relying on people to travel to a drop-off point or what we're currently faced with which is a whole lot of rubbish and not a lot of action.

  • Dr Wanchun Liu is a researcher in the Centre for Internet of Things and Telecommunications at the University of Sydney. She currently researches AI-driven circular economy technology.

Listen to the Fuzzy Logic Science Show at 11am Sundays on 2XX 98.3FM.

Send your questions to AskFuzzy@Zoho.com Twitter@FuzzyLogicSci

We've made it a whole lot easier for you to have your say. Our new comment platform requires only one log-in to access articles and to join the discussion on The Canberra Times website. Find out how to register so you can enjoy civil, friendly and engaging discussions. See our moderation policy here.

Sign up to read this article
Read news from 100’s of titles, curated specifically for you.
Already a member? Sign in here
Related Stories
Top stories on inkl right now
One subscription that gives you access to news from hundreds of sites
Already a member? Sign in here
Our Picks
Fourteen days free
Download the app
One app. One membership.
100+ trusted global sources.