Roboverse Reply will be spearheading the EU-funded project “Fluently.” The primary objective of this initiative is to create a platform fostering genuine social collaboration between humans and robots in industrial settings by incorporating the latest AI-driven decision-making advancements.
Set to span three years, the project endeavors to produce a platform and a portable device for industrial workers and robots. This will allow machines to more accurately interpret voice, content, tone, and translate gestures directly into robotic commands. Additionally, the project envisions the establishment of a training center named “Fluently RoboGym”. Here, industrial workers and robots can practice seamless interactions during production processes.
Anticipated practical applications for human-robot collaboration target critical future value chains in European industries. Examples include tasks requiring physical exertion, yet also demanding human expertise, such as the disassembly and recycling of lithium cell batteries, maintenance processes in aviation, and refurbishment of intricate industrial parts using additive manufacturing.
Twenty-two partners, including the Swiss university SUPSI, are involved in the project. Anna Valente, head of the SUPSI Laboratory for Automation, Robotics & Machines and a member of the Swiss Science Council, elaborated, “Fluently aims to train robots to become team members that optimally support human workers. As scientific and technical coordinators, we’ve shaped Fluently to be a pivotal step forward in advanced human-robot collaboration, fostering best practices and creating more inclusive and interactive ecosystems.”
Having successfully concluded its first developmental year, the project has met initial milestones. Focus now shifts to three primary work packages:
- Design of the Fluently unit, entailing device design, software testing, and integration into wearable straps and robotic systems.
- Development of AI models, covering architectural design, edge computing, Robo-Gym model training, and support for human-robot teamwork.
- Design and establishment of Robo-Gym, involving the setting of specifications, objectives, and the construction of three training zones.
The Fluently system will employ cutting-edge technologies to ensure seamless human-robot communication. Natural language processing, hands-free remote collaboration tools, physiological signal monitoring, and eye-tracking are among the subjects to be explored and integrated.
Filippo Rizzante, CTO of Reply, commented, “We’re honored to coordinate the trailblazing Fluently project. Uniting academic and industry partners, our goal is to craft an empathetic robotic platform capable of interpreting speech content, tone, and gestures, making industrial robots accessible to individuals of all skill levels. Robots equipped with Fluently will continually support humans’ physical and cognitive tasks and learn from, as well as gain experience with, their human counterparts.”