Live· ·Issue N°—
CO₂— ppm·Temp anomaly—°C·CH₄— ppb

Redwood Materials Deploys AI Battery Recycling System With Over 98% Sorting Accuracy

Redwood Materials Deploys AI Battery Recycling System With Over 98% Sorting Accuracy

US battery recycling company Redwood Materials has developed and deployed an automated battery-sorting system that uses computer vision and machine learning to identify different battery chemistries with over 98% accuracy. The system is already processing battery waste collected across the United States, helping improve material separation before recycling. The technology supports Redwood's industrial-scale recycling operations, including a rotary calciner designed to process more than 40,000 metric tonnes of batteries annually, equivalent to approximately 15–20 gigawatt-hours (GWh) of battery capacity.

 

The automated sorting line processes mixed batteries from consumer electronics, electric vehicles and other sources. Batteries are separated into a single-file stream before passing through a sensor tunnel equipped with high-resolution cameras, a laser profilometer and weighing equipment. These systems capture information about battery labels, chemical markings, physical dimensions, and weight, which machine learning models analyze on an on-site GPU computing system. The technology identifies battery chemistry in real time and directs each unit towards the appropriate recycling stream, operating at a throughput of several metric tonnes per hour.

 

Read more: Aramco Ventures Invests in ROTOBOOST's Low-Carbon Hydrogen Technology

 

Redwood developed the system to address challenges associated with manual sorting, where differences in battery chemistry can affect processing efficiency and the quality of recovered material. Incorrect classification can introduce unwanted materials into recycling equipment, while rejecting suitable batteries can reduce the recovery of valuable metals such as lithium, nickel and cobalt. The company therefore tracks both types of classification errors separately, using operational data and human verification to continuously improve its machine learning models.

 

An earlier version of the system used X-ray technology, but Redwood shifted towards conventional cameras after determining that external battery markings and physical characteristics provided sufficient information for classification. The newer approach reduces equipment complexity and operating costs while improving scalability. Each sorting decision is recorded alongside its supporting sensor data, enabling engineers to investigate downstream processing issues and retrain the models using difficult or incorrectly classified examples.

 

Explore OneStop ESG Marketplace: Circular economy & recycling

 

Redwood says its internal engineering team developed and brought the automated line into operation within several months. The technology is intended to improve feedstock consistency, reduce contamination and support greater recovery of critical battery materials for reuse in domestic manufacturing. Although the company reports sorting accuracy exceeding 98%, it has not disclosed independently verified improvements in material recovery rates, processing costs, or greenhouse gas emissions attributable specifically to the new sorting system.

 

Source: Redwood Materials

 

Subscribe to our newsletter for more insights, case studies, and ESG intelligence.

 

Explore ESG Solutions on our marketplace - OneStop ESG Marketplace.

 

Keep abreast of the top ESG Events on OneStop ESG Events.

 

OneStop ESG Educate: Your go-to source for top ESG courses and training programs tailored to your needs.

 

Stay informed with the latest insights on OneStop ESG News.

 

Discover meaningful career opportunities on OneStop ESG Jobs.

Comments

Have a thought on this? Share it with other readers.

Got something to say? Sign in to join the discussion.

Recommended Reads

Have a sustainability story to share?

If you’re working on ESG, climate action, governance, social impact, or sustainable innovation, your perspective matters.

Publish articles, insights, case studies, or thought leadership and reach a global sustainability audience.

Open to professionals, researchers, founders, and practitioners.