If you are a supermarket chain dealing with the labor-intensive task of restocking shelves with diverse products — this project developed a bimanual mobile robot that can handle objects of various sizes and materials to automate inventory placement.
Human-like Two-Armed Mobile Robots for Flexible Object Handling in Unstructured Environments
Imagine a robot that can pick up things just like a person does, using two arms to balance heavy or oddly shaped items. Instead of needing a perfectly clean factory floor, it uses a mix of sight and touch to figure out what an object is and how to move it. It can even push or slide things across a table if they are too heavy to lift, making it much more versatile than a standard robotic arm.
What needed solving
Current industrial robots require controlled settings and pre-programmed tasks, making them unable to handle unknown, heavy, or deformable objects in real-world, human-populated spaces.
What was built
A bimanual mobile robotic platform with multi-modal sensing (vision, touch, proximity) and a web portal for system interaction.
Who needs this
Who can put this to work
If you are an airport service provider dealing with heavy or irregularly shaped luggage — this project developed non-prehensile manipulation (pushing/dragging) that allows robots to move baggage efficiently even when it exceeds standard payload limits.
If you are a food producer dealing with deformable or unknown objects in a cluttered space — this project developed a multi-modal sensing system using vision and touch to safely manipulate items without needing a pre-programmed model.
Quick answers
What is the cost or price of implementing this system?
Based on available project data, specific pricing or implementation costs are not provided.
Can this be scaled to an industrial level?
The project is testing the system in 3 pilot sites at TRL5, specifically targeting retail and transport environments to prove its industrial viability.
How is the intellectual property or licensing handled?
Based on available project data, the specific IP and licensing terms are not disclosed in the summary.
How does the robot handle objects it has never seen before?
It uses an adaptive fusion of vision, proximity, and tactile sensing to recognize and manipulate objects that are not known in advance or modeled.
When will the technology be ready for full commercial use?
The project period runs until 2027-04-30, with current goals focused on reaching TRL5 in pilot sites.
Who built it
The consortium is well-balanced for technology transfer, consisting of 13 partners across 7 countries. With a 31% industry ratio (4 companies), the project has a strong commercial anchor, while the 6 universities and 3 research centers provide the necessary deep-tech expertise in mechatronics and sensors.
Contact ETHNIKO KENTRO EREVNAS KAI TECHNOLOGIKIS ANAPTYXIS in Greece
Talk to the team behind this work.
Contact us to explore licensing opportunities for bimanual robotic control.