• Vision
• Research Areas
• Laboratory
• Events
• Admission
• People
• Publications
• Access
• Link

Reserch Areas

What is Swarm Robotics?

　Swarm Robotics (SR) is a field of research on the coordination of large numbers of robots that has emerged from the observation of these social creatures. Robotic systems based on SR are expected to have the following three characteristics　Swarm Robotics (SR) is a field of research on the coordination of many robots that has emerged from the observation of these social creatures. Robotic systems based on SR are expected to have the following three characteristics.

• Robustness
  The system can function as a system even if a small number of robots are missing, e.g. due to failure
• Flexibility
  Can function as a system in different environments
• Scalability
  The system can function as a system even if the number of robots is varied

In SR, ‘how to acquire robot behaviour’ is one of the central issues and has been widely studied. In our laboratory, we aim at acquiring autonomous robot behaviour through Evolutionary Robotics and Reinforcement Learning. We aim at acquiring autonomous robot behaviour through Evolutionary Robotics and Reinforcement Learning. We also design models of robot behaviour using Biomimetics, which can be used for the acquisition of complex robot behaviour. Complex robot behaviour is acquired by designing robot behaviour models using Biomimetics.
　On the other hand, not much research has been done on ‘how to analyse acquired robot swarms’. Therefore, this laboratory attempts to analyse robot swarms based on complex network theory as an approach for analysing robot swarms.

Approaches to herd behaviour generation

• Evolutionary Robotics
• Reinforcement Learning
• Biomimetics

Behavioural analysis approach

• Clustering

• ▲TOP