Robot control - RITMO

How is rhythm processed in the human brain, and how can we model rhythm in machines? These are central research questions at the new RITMO Centre of Excellence.

We aim to take inspiration from rhythm in humans and other biological systems and develop models of rhythmic motion which can be applied to robotic and computing systems.

Bildet kan inneholde: linje.

Possible topics for MSc projects within RITMO could be:

  • Central pattern generators (CPGs) as control mechanisms for robots. CPGs are biological neural networks that produce rhythmic patterned output, and these have been used as inspiration for robotic control systems. We would like to  explore the conditions for synchronization of robot motion to a rhythmic input (such as music). This will be done with a simulated robot using a newly developed robot controller based on a non-linear CPG. Possible questions are in what cases synchronization can handle gait transitions (e.g. walk to trot) or how it competes with sensory feedback.
  • Explore rhythmic aspects in collective systems. These can be systems consisting of heterogenous agents, which have limited and different knowledge. In particular, we have earlier investigated a collective of musical agents, demonstrated in the SoloJam application. We would like to investigate this further with new types of agents and control mechanisms. One approach to this could be to simulate a robotic swarm and study their interactions. It could also be possible to study interactions on the Dr. Squiggles and self-playing guitars platforms (see links on the right).
  • Explore how rhythm is processed by the human brain. This task would involve investigating brain data (e.g. from EEG or fMRI signals) with machine learning methods and see if it is possible to correlate and/or reproduce parts or events from the input signal. For example, would it be possible to reconstruct a musical beat that a human has listened to?
  • Mapping movements and other input data from humans in performances to robots. This could be robotic swarms or other types of robots, like a robotic arm. One challenge here would be how to best transform human performance to robotic movements taking place on another "body" with different dynamics.

The RITMO - Centre for Interdisciplinary Studies of Rhythm, Time and Motion started in 2018. Researchers in the network come from the ROBIN group (Informatics), FRONT Neurolab (Psychology), and the Department of Musicology, and have access to state-of-the-art laboratories.

 

 

Emneord: robotics, machine learning, self-aware systems, ritmo
Publisert 15. okt. 2020 14:20 - Sist endret 17. okt. 2021 22:56

Veileder(e)

Omfang (studiepoeng)

60