Subscribe to our newsletter to receive the latest news and events from TWI:

Subscribe >
Skip to content

MANiBOT Promises a Step Change in Bi-manual Mobile Robots

Fri, 16 February, 2024

The EU-funded MANiBOT project, aiming to empower service robots with superior physical capabilities, was launched at the Centre for Research and Technology Hellas (CERTH) in Thessaloniki, Greece, in late 2023.

The consortium is led by research institute CERTH, and its expertise is complemented by the Central Institute for Labour Protection - National Research Institute and six universities: Aristotle University of Thessaloniki, Technical University of Darmstadt, Technical University of Vienna, University of Bristol, University of Burgos and Sant'Anna School of Advanced Studies, along with leading industry player Asea Brown Boveri SA, and commercial partners Fraport Regional Airports of Greece Management, Diamantis Masoutis Supermarkets and Schwarz Digital GmbH & Co. KG.

MANiBOT aims to equip mobile service robots with the capacity to handle a wide variety of unknown objects in a human-like way in environments where people are also present. This will unlock the true potential of these robots, making them attractive to new, major industry and service sectors.

Despite advances in robotic perception, understanding and control, collaborative service robots still have limited physical performance compared to that of humans. Industrial-grade robots demonstrate strength, dexterity and speed but only in the context of handling well-known objects in controlled environments.

Dr. Dimitrios Giakoumis from CERTH, the Project Coordinator, explains "The MANiBOT vision is of service robots capable of manipulating diverse, and not necessarily well-known, objects efficiently, in a human-like manner". "To achieve this, MANiBOT aims to advance individual technologies for robot perception, cognition and bi-manual manipulation, as well as their coupling", he added.

The project seeks to revolutionise the robotics landscape by enhancing robots' handling skills, including simple grasping, pick-and-place operations, bi-manual and non-prehensile manipulation, and ensuring adaptive responses to changing environments or the properties of objects. To achieve these capabilities, innovations will be developed in the fields of advanced environment understanding, efficient manipulation techniques, robot cognitive functions and physical intelligence. The researchers will implement their solutions with a focus on baggage handling and supermarket shelves restocking by piloting the robots in relevant environments.

TWI Hellas is proud to be part of the MANiBOT consortium since it brings together the multidisciplinary expertise of key players from seven countries in the research of robotic technologies. More specifically, the team is contributing to the project by focusing on federated explainable learning, developing the MANiBOT solution technical specification and architecture, intuitive and responsive human-robotic interaction (HRI), and human-centric robot scheduling, as well as the robot operating system (ROS)-based MANiBOT demonstrator system integration, and lab and pilot sites' preparation, amongst other tasks.

Moreover, lead Software Engineer Mike Karamousadakis points out that TWI Hellas will investigate novel solutions and algorithms pertaining to federated explainable learning. "The issue of federation is a major one in robotics as robots are usually operating in commercial and/or industrial spaces. Collective learning based on data from the robots' sensors is challenging because industries don't want this data to be shared due to reasons including organisational policies and privacy concerns. Federated learning fills this void with techniques for distributed learning based on the sharing of models instead of data. In addition, the machine learning (ML) algorithms used in robot perception and/or task planning are difficult for humans to understand; thus, there is an issue of trust between robots and their operators. Explainable learning comes to the rescue by providing new algorithms and techniques for the models that are used for the interpretation, thereby leading to increased trust between humans and robots."

For more information please email: