Will AI Be The Answer To The World’s Recycling Crisis?

Recycling topic is gloomy. But there is hope.

A photo by Lisa Fotios via Pexels

It’s 2020. What happens to our rubbish?

Although recycling nowadays is an absolute buzz-word, only 9% of all plastic is actually recycled. The rest of it usually fills up the oceans forming trash islands. According to statistics, in 30 years, the amount of plastic (only) in landfills will be 35,000 times heavier than the Empire State Building.

The recycling crisis is undeniable.

Two years ago, in 2018, China restricted its import of recyclable waste products from the USA. This event, together with the recent technological developments in AI and robotics, has brought a significant interest of some western companies towards using AI as a waste sorting and recycling tool.

Is AI the Answer for the Recycling Crisis?

Photo by Possessed Photography on Unsplash


Sorting rubbish is one of the most monotonous, repetitive, low-paid and even dangerous jobs, that automation can cover. In my previous article, I am writing about the reasons why humans should not be afraid of automation. And recycling and waste sorting are one of the illustrative examples of that.

AI-enabled robots are usually a cost-efficient and a few times more productive than humans. What is more, robots would not threaten their life and health by encountering with a hypodermic needle or other unsanitary, sharp and potentially dangerous stuff. What is more, the growing demand for AI systems in recycling centres will produce a set of new roles and further expand the job market.

I would not cover the financial side of the recycling market, but this is an important note that supports the idea that AI will lead the recycling revolution. It is estimated that already by 2024, waste sorting and recycling AI robots market will be worth $12 billion.

Photo by Artem Beliaikin on Unsplash

How Does AI sort waste?

With computer vision, AI-robots can identify logos, colours, shapes and textures and proceed with sorting based on this data. For instance, AI can recognize an Oreo logo, blue-coloured package and a picture of a cookie. Based on this info it can ‘assume’ it’s a package from the plastic cookie pack.

Photo by Arseny Togulev on Unsplash

Similar to any AI-enabled robot, firstly AI-specialists need to ‘teach’ the system with the required knowledge. Based on that, AI processes millions of images of waste categories and subcategories, packages, logos, shapes, colours, textures, etc. In the sorting process, the system continually teaches itself to recognize the most obvious attributes of different waste types. Whether the waste is smashed, broken or, for instance, dirty, AI teaches itself to identify these attributes no matter the condition it’s in.

There are many different designs of AI-powered robots that can be enabled by various systems to identify the waste types. Some of these robots are designed to scan an object with computer vision or physically grab the particular object. In this case, its’ sensors would identify the object’s characteristics, height, weight, material, etc. Apart from computer vision, waste sorting AI and robots can be enabled with spectroscopy, laser sensors, metal detectors or simply, magnets to capture metal objects from mixed waste.

Is AI already helping us to reach sustainability?

Recycling and trash sorting robots are already starting to raise interest and gain traction. Even before Covid-19 struck.

For instance, Sweden is famous for many things. Apart from its gorgeous national parks, breathtaking landscapes and minimalistic Swedish design, Sweden is now famous for being one of the most sustainable countries with the recycling rate of 99%. Swedish recycling revolution is a product of reshaping the mind of the nation to more environmentally-aware, the massive work of government and technology. AI-enabled waste scrapping robots are already supporting Sweden in the recycling revolution and helping the country become almost fully sustainable.

A photo by ready made via Pexels



Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store