Norwegian version of this page

Synchronized Robotics (completed)

The project will develop Collective Robotics through Real-time Entrainment of Evolved dynamical Systems.

Two attractors in purple and red on a blue background. Illustration.

Dynamical systems often settle into cycles, but these can still be highly complex. Photo: Lorenz attractor, Wikimedia Commons CC 4.0

About the project

Humans and many other vertebrates contain oscillatory networks known as central pattern generators that generate movement patterns. These networks are able to entrain to rhythms and adapt to social information. However, we still do not understand the mechanisms behind this very well.

The project “COllective ROBOtics through Real-time Entrainment of Evolved dynamical Systems” (COROBOREES) uses evolutionary methods to create complex non-linear dynamical systems that allow flexibility while retaining stability. These will be used as central pattern generators with tunable social responsiveness in legged robots.

Purpose

Using this generative approach, in both simulated and prototyped robots, the project will investigate:

  • Minimum requirements for rhythmic entrainment
  • How rhythmic complexity affects entrainment
  • Conditions for synchronization between multiple agents

The results of this research will also inform the improvement of adaptive motion in robots, as well as provide a testbed for controllable studies of collective behaviour.

Duration

01.09.21 - 07.02.24

EU-flagget.

Financing

This project has received funding from the European Union’s Horizon 2020 research and innovation programme under the Marie Sklodowska-Curie grant agreement No 101030688.

Selected publications

  • Szorkovszky, A., Veenstra, F., & Glette, K. (2022). Rapid rhythmic entrainment in bio-inspired central pattern generators. In 2022 International Joint Conference on Neural Networks (IJCNN). IEEE. link
  • Szorkovszky, A., Veenstra, F., & Glette, K. (2023). Central pattern generators evolved for real-time adaptation to rhythmic stimuli. Bioinspiration & Biomimetics 18 046020. link
  • Szorkovszky, A., Veenstra, F., Lartillot, O., Jensenius, A.R. & Glette K. (2023). Embodied tempo tracking with a virtual quadruped. In Proceedings of the Sound and Music Conference 2023, pp 283-288. link
  • Szorkovszky, A., Veenstra, F. and Glette, K. (2023). From real-time adaptation to social learning in robot ecosystems. Frontiers in Robotics and AI, 10 1232708. link
Published Dec. 15, 2021 9:11 AM - Last modified Mar. 10, 2024 8:58 AM

Contact

Alexander Szorkovszky
Head of project

Participants

Detailed list of participants