Tomra Recycling, a global supplier of sensor-based sorting technology, announces that it will strengthen its position in AI solutions for recycling at Ifat 2026, held from 4 to 8 May in Munich. In hall B6, stand 339, the company will highlight the 30th anniversary of the Autosort system and present the development of its technologies from early rule-based optical systems to today's fully integrated solutions. Under the slogan "30 years of AI sorting", Tomra also plans to launch new deep learning and artificial intelligence-based solutions.
From rule-based sorting to full integration and connectivity
Since the first version of the Autosort system was introduced 30 years ago, Tomra has helped drive the industry's transition from pioneering automated sorting to fully integrated, connected systems. Visitors to the company's stand will be able to explore the evolution of Autosort and the broader history of sensor- and AI-based sorting, illustrated by successive generations of the system introduced to the market since 1996.
The starting point was the 1996 edition, which brought high-capacity automated sorting based on smart optics and early artificial intelligence principles. In the following years, the use of hyperspectral imaging and greater software flexibility made it possible to adapt systems to new material streams. Optical upgrades improved illumination quality and enabled the detection of black materials, while advanced AI-based "material intelligence" expanded the range of fractions that could be distinguished and increased sorting accuracy.
An important milestone was the introduction in 2019 of GAINnext technology, described as the first deep learning solution for waste sorting on the market. As an add-on module to Autosort, GAINnext enabled completely new applications and helped solve previously challenging tasks, including sorting plastics intended for food-contact applications.
Today, Autosort is described as one of the most advanced sorting systems on the market. It combines artificial intelligence, multi-sensor technology and advanced image processing with functionalities such as GAINnext, waste stream analysis and other digital tools for real-time monitoring, data-based optimisation and smooth plant integration. This makes it a digitally connected core of many recycling plants focused on high-quality output fractions.
Intelligent, connected and user-friendly ecosystems
Lars Enge, EVP and Head of Tomra Recycling, emphasises the importance of the anniversary: "For three decades, the Autosort brand has set the benchmark for performance in the recycling industry. Ifat 2026 offers us the ideal stage not only to reflect on our history, but also to share a clear vision of what comes next. We will introduce several new, impressive deep learning and AI solutions that help bridge the gap to true circularity. These innovations enable sorting plants to move beyond high-speed detection toward truly intelligent, interconnected and user-friendly ecosystems."
Next-generation sorter control and analytics
Alongside the Autosort system and its award-winning GAINnext technology, one of the main elements of Tomra's presentation at Ifat 2026 will be a new, intuitive software platform. The Tomra Local Control solution allows operators to manage all sorting machines from a single interface within the plant network. The platform provides full visibility of every machine on the line, delivering updates every second on operating status and performance trends.
Operators can remotely adjust configurations, activate sorting plans and create custom programs from the office. This has a significant impact on plant operation, turning plant management into a faster, more efficient and more data-driven process.
The Tomra stand will also host the team from PolyPerception, a provider of AI-based waste analysis technology. This solution adds an additional layer of "intelligence" to modern sorting plants. The deep learning waste stream analyser provides full transparency and traceability of operating parameters across the entire line, supplying real-time data on material composition, recovery and purity of fractions.
By turning continuous stream data into reliable, actionable information, PolyPerception enables operators to increase efficiency, optimise plant performance and make data-driven decisions, thereby reinforcing Tomra's vision of a smarter, more connected and future-ready recycling ecosystem.
Launch of new deep learning and AI solutions
At Ifat 2026, Tomra also plans to launch a range of new deep learning and AI-based solutions that will be presented to the wider public for the first time. Company experts will be available at the stand to discuss the latest developments in deep learning as well as the broader portfolio of sensor-based sorting technologies.
Enge concludes the announcement of the company's participation in the event with the following words: "We welcome customers, partners and industry companions to our Tomra booth in Hall B6 to discover what’s next in AI sorting."
Tomra invites visitors to meet the company at Ifat 2026 in hall B6, stand 339. Detailed information on the technology solutions to be showcased is expected to be provided in materials accompanying the event.
Tomra Autosort system with Gainnext module presented in the context of 30 years of AI-based sorting
