AI firm secures Amazon funds to transform mixed waste recycling with robots

07 Mar 2024 --- Glacier has secured US$7.7 million in funding to advance its development of AI-driven waste sorting robots for enhanced recycling efficiencies and data collection. Co-founder Rebecca Hu highlights the robots’ capability to identify and sort recyclables, emphasizing the technology’s potential to reduce waste by improving the purity and recovery rate of recycling streams.

The company has raised the funding from New Enterprise Associates (NEA) and Amazon’s Climate Pledge Fund. Additional investors include AlleyCorp, Overture Climate VC and VSC Ventures. 

The robotics company intends to utilize the funding to bolster its recycling capabilities, particularly in light of its recent collaboration with Amazon. This partnership aims to enhance the traceability and recovery processes of recyclable materials.

AI-enhanced recycling 
An estimated 80% of residential recyclables in the US are not recycled, contributing to environmental pollution and resource depletion. Glacier asserts that traditional recycling facilities need more access to the highly specialized machinery required for efficiently sorting mixed materials. 

Hu tells Packaging Insights that the company’s robots are installed in recycling facilities to sort and decontaminate recyclables that can then be turned into new feedstock.

Glacier’s technology can identify anything a human can.“We help facilities rescue more recyclables from landfills and pick out contaminants that reduce the purity rate of their recycling streams,” she says.

“Using our proprietary AI model, Glacier robots can identify over 30 different categories of items, spanning from broad categories like PET bottles and aluminum cans to specific items such as toothpaste tubes and cat food cans.”

“Glacier robots can also effectively sort a wide array of items, including challenging materials such as grocery bags and trash bags, which are notoriously difficult to sort using traditional recycling processes.”

Human-sized robots
Glacier’s technology is made for easy deployability. Space limitations, installation downtime and capital requirements often constrain recycling facilities. To address these concerns, Glacier designed a robot that occupies the same footprint as a person and can be installed with no facility downtime and no heavy machinery.

“One Glacier robot can pay back in less than one year,” says Hu.

“We designed our purpose-built robot considering how space-constrained recycling facilities are — many facilities are eager to install robots but don’t have space for large machinery. As a result, we designed our robot to occupy as little as 3 feet of conveyor length — the same amount of space as a person takes up, and only one-third the size of some similar equipment doing the same job.” Glacier robots can identify over 30 different packaging categories.

Glacier’s recycling tech 
Glacier develops “best-in-class” computer vision to identify high-value items in the waste stream, asserts Hu. 

“Our cutting-edge AI can accurately identify anything a human can. Not only does this AI deliver valuable insights to recyclers, brands and governments — it also powers our purpose-built sorting robot that automates the ability to divert recyclables from landfills and purify recycling streams.”

When asked about cost challenges, Hu says that many Glacier robots have been added to existing recycling facilities.

“We built our technology to be as plug-and-play as possible. As a result, our robots can often be installed on an existing line with no facility downtime, major retrofits or heavy machinery required,” she says.

By Natalie Schwertheim