Intelligent robots that make logistics more efficient and customer service more comprehensive in stationary retail – this was the idea behind the REFILLS research project launched by the University of Bremen in 2017. Three years later, this led to the founding of the start-up Ubica Robotics, which is successfully putting the scientific results into practice. Together with dmTECH GmbH – the IT subsidiary of the drugstore chain dm-drogerie markt – Ubica and the Institute for Artificial Intelligence at the University of Bremen (IAI) have now been awarded the prestigious Technology Transfer Award 2022 at the European Robotics Forum in Rotterdam.
Latest Generation of Robots Navigates Independently
The latest generation of Ubica robots is capable of autonomously navigating through a retail store while recognizing the inventory of products on the shelves. Based on this scan, a so-called digital twin of the store is generated, i.e., a virtual version that provides a wealth of important information on the operation of the store – from replenishment requirements and the best product placement to the ideal packing of pallets in the central warehouse in order to make shelf replenishment in the store as efficient and trouble-free as possible for customers.
“Ubica technologies can become a disruptive innovation in the field of intelligent robotics,” emphasizes Professor Michael Beetz, head of the Institute for Artificial Intelligence (IAI) at the University of Bremen. With the IAI and the Collaborative Research Center EASE, which is also headed by Professor Beetz, the university is one of the world’s leading research institutes at the interface of artificial intelligence and robotics. “The euRobotics Award is a wonderful affirmation that the transfer of our research results to industry is successful and also attracts international attention,” says Beetz.
Digital Twins: Generating a Virtual Environment
Ubica’s autonomous scanning robots use state-of-the-art artificial intelligence and machine learning to generate the virtual environment in the form of digital twins. “What’s especially unique about this is that the robots are able to understand these digital models in detail and link them to background knowledge,” explains Beetz. This capability is made possible by a combination of semantic models and data-driven learning algorithms.
So far, no comparable technologies exist in stationary retail. “Our robots enable a significant improvement in a store’s internal processes,” explains Georg Bartels, Cofounder and Chief Technology Officer of Ubica. “Retailers get a detailed and fully data-integrated view of their stores, which allows for targeted support of in-store merchandise flows.” In addition, the assortment can be better optimized to meet the needs of customers and arranged in the store. The connection of shopping apps is also a conceivable prospect in the future.
“With the help of machine learning and big data, Ubica offers the opportunity to develop innovative service offerings for customers,” says Jan-David Walter of dm subsidiary dmTECH. “For example, it offers the possibility of individual product recommendations or the connection of delivery services. We want to offer many of the benefits of online stores to brick-and-mortar customers as well.”
Professor Michael Beetz
Institute for Artificial Intelligence
Faculty of Mathematics and Computer Science
University of Bremen
Phone +49 421 218-64000