Central pattern generators for locomotion control in animals and robots: a review

Area of Science:

• Neuroscience and Robotics
• Computational Neuroscience
• Biomechanical Engineering

Background:

• Locomotion control is a complex problem where neuroscience and robotics intersect.
• Central Pattern Generators (CPGs) are neural circuits producing rhythmic outputs for coordinated movements with minimal input.
• Research on CPGs spans neurobiology, computational modeling, and robotic applications.

Purpose of the Study:

• To review research on locomotor CPGs, bridging neurobiological findings and robotic implementations.
• To explore the use of CPG models in controlling articulated robots.
• To discuss the reciprocal relationship between robotics and CPG research in understanding biological systems.

Main Methods:

• Review of neurobiological studies on vertebrate locomotor CPGs.
• Analysis of numerical modeling and computational approaches to CPGs.
• Examination of CPG models (neural networks, coupled oscillators) applied to robotic locomotion.

Main Results:

• CPG models can effectively control articulated robots, enabling diverse locomotion modes.
• Robots serve as valuable tools for investigating the functional mechanisms of biological CPGs.
• Various design methodologies for CPGs in robotics have been explored.

Conclusions:

• The interaction between neuroscience and robotics offers significant advancements in understanding and replicating locomotion.
• CPG models provide a robust framework for robotic control, with ongoing research into their design and application.
• Future research directions lie in both refining biological CPG understanding and enhancing robotic locomotion capabilities.