Humanoid developer Neura acquires cleaning automation company Adlatus Robotics
Neura Robotics is expanding its physical AI ecosystem into another key application area. Neura Mobile Robots is acquiring 100 percent of the shares in Adlatus Robotics, an established provider of autonomous cleaning robotics.
Following the recent addition of Bosch Rexroth’s Active Shuttle to its portfolio, this marks another step in Neura’s strategy: bringing together different robots and applications on a shared physical AI infrastructure, turning intelligence from individual machines into a scalable platform.
Adlatus has spent years developing autonomous cleaning and sweeping robots for industry, logistics, healthcare, and commercial and public spaces. The company has hundreds of systems in the field, its own navigation software, and extensive knowledge of real-world cleaning processes.
From autonomous cleaning robot to intelligent physical AI system
The acquisition is designed to build on this foundation. The goal is to equip Adlatus systems with additional sensor and AI capabilities going forward and to gradually connect them to the Neuraverse.
Over time, a cleaning robot should no longer simply follow a predetermined route, but increasingly understand its environment: recognizing different surfaces and types of dirt, interpreting situations, selecting appropriate cleaning strategies, and learning from real-world deployments.
The Neuraverse makes it possible to bring together different AI models, sensor data, and capabilities to enable this. Step by step, a specialized autonomous device becomes an intelligent agent within a larger physical AI ecosystem.
David Reger, founder and CEO of Neura Robotics, says: “We’re not buying Adlatus just to add another cleaning robot to our portfolio. We want to give machines like this a new brain. Cleaning is a massive global labor market and, at the same time, a perfect example of what physical AI can transform.
“When a robot doesn’t just navigate its environment but understands what it sees, senses, and needs to do, an entirely new generation of machines emerges. That’s exactly what we’re building the Neuraverse for.”
An open platform instead of a closed ecosystem
Integrating Adlatus is explicitly not a move away from collaboration with other manufacturers. Neura’s goal is to establish the Neuraverse as an open physical AI infrastructure for a wide range of robots, manufacturers, and form factors.
Manufacturers of cleaning robots, mobile robots, and other autonomous systems should eventually have the opportunity to connect their machines to technologies, AI models, and capabilities from the Neura ecosystem as well. Adlatus thus becomes both a product platform and a real-world testbed for advancing physical AI.
“Physical AI will only truly scale if it becomes an ecosystem,” says David Reger.
Industrial know-how meets AI and scale
Going forward, Adlatus will benefit from the Neura Group’s manufacturing expertise, international sales structure, and technology platform.
At the same time, the Adlatus team brings deep industry knowledge, established customer relationships, and years of experience from real-world cleaning operations.
This combination is exactly what makes physical AI possible. Unlike classic generative AI, physical AI depends on data and experience from the real world. Machines have to learn how physical tasks are actually carried out.
Adlatus’s installed base provides a valuable foundation for developing, validating, and ultimately scaling new capabilities under real-world conditions.
Andreas Lindemann, managing director of Neura Mobile Robots, adds: “Our ambition is to take mobile robotics beyond classic material transport. Autonomous cleaning is a logical next step in that direction.
“Looking ahead, we also plan to integrate these solutions into our Fleet Manager, opening up additional applications for existing and new customers alike.
“With Adlatus, we’re laying the groundwork to grow this business deliberately and, over the long term, establish it as its own product family within our AMR portfolio.”
Another building block for a European physical AI industry
With its recent portfolio expansions, Neura is pursuing a broader industrial strategy. Europe holds decades of expertise in mechanical engineering, automation, sensor technology, and industrial processes.
Neura aims to combine that expertise with modern artificial intelligence to build a new generation of intelligent machines out of Europe.
The goal isn’t to replace successful European technologies, but to connect them with one another and advance them for the next technological generation.
“Europe has an incredible amount of industrial knowledge. We don’t need to reinvent everything – we need to combine the best of what already exists with artificial intelligence and scale it much faster,” Reger says.
“That’s exactly what we’re doing. Companies like Adlatus bring decades of experience from the physical world. Neura brings the physical AI infrastructure. When we combine these strengths, we can build an entirely new industry out of Europe.”
A partnership already proven in practice
Neura Robotics and Adlatus have been collaborating under a strategic partnership since September 2025. Their first joint public appearance took place at CMS in Berlin. The two companies are also aligned culturally in their shared understanding of robotics.
“Adlatus” is Latin for “helper”. Since its founding, the company has been committed to developing machines that support people in everyday and physically demanding tasks.
This ties directly into Neura Robotics’ founding mission: “We serve humanity.”
Robots are not meant to replace people for their own sake, but to take on dangerous, monotonous, and physically demanding work – freeing up new space for human capability.
Main image (from left): Adlatus’ Stefan Erhardt; Neura Robotics’ David Reger; and Neura Mobile Robots’ Andreas Lindemann
How it works
Once you click Generate, Ollama reads this article and crafts 5 comprehension questions. Your answers are graded against the article content — general knowledge won't be enough. Score 70+ to count toward your certificate.
Questions are cached — you'll always get the same 5 for this article.