Hebbian Plasticity in CPG Controllers Facilitates Self-Synchronization for Human-Robot Handshaking.

Area of Science:

• Robotics
• Computational Neuroscience
• Human-Robot Interaction

Background:

• Human social interactions often lead to unconscious synchrony, especially in rhythmic movements.
• Modeling synchronized movements is crucial for natural human-robot interaction.

Purpose of the Study:

• To propose a model of plastic neural controllers for emergent synchronized movements in physical interactions.
• To investigate the application of this model to robotic handshaking.

Main Methods:

• Developed a controller using central pattern generators (CPGs) with neuronal and synaptic Hebbian plasticity.
• Implemented plastic CPGs in a simulated robotic arm for handshaking tasks.
• Trained the robot to adapt to external forces with varying frequency and amplitude.

Main Results:

• The plastic CPGs successfully achieved autonomous synchronization between the robotic arm and external forces.
• Hebbian plasticity enabled natural coordination even with changing movement frequencies.
• Synchronized movements significantly reduced the robot's actuator energy consumption.

Conclusions:

• The proposed model effectively generates synchronized and coordinated movements in physical human-robot interactions.
• Neuronal and synaptic plasticity are key to adaptive and energy-efficient robotic behavior.
• This approach enhances the naturalness and efficiency of robotic handshakes.