A bio-inspired customizable mechanical central pattern generator enables one-to-many scalable pneumatic control

Area of Science:

• Robotics and Automation
• Mechanical Engineering
• Bioinspired Design

Background:

• Pneumatic robotics with multiple actuators face challenges with component redundancy and complex tubing.
• Conventional control systems require one valve per actuator, limiting scalability and efficiency.

Purpose of the Study:

• To develop an electronics-free, scalable pneumatic control core for multi-actuator systems.
• To propose a bioinspired mechanical central pattern generator (CPG) for one-to-many pneumatic control.

Main Methods:

• Designed and characterized a customizable mechanical CPG analog integrating multi-channel timing logic.
• Utilized discrete and mechanically-latched state transitions for robust control.
• Experimentally validated performance with four- and five-channel configurations under varying conditions.

Main Results:

• Achieved efficient 1:n pneumatic control, driving multiple actuators from a single input.
• Demonstrated pressure retention rates exceeding 90.8% across different pressures and loads.
• Confirmed exceptional sequence fidelity and robustness against physical disturbances.

Conclusions:

• The mechanical CPG offers a scalable, electronics-free solution for complex pneumatic systems.
• This approach fundamentally reduces controller quantity and tubing complexity.
• Demonstrated versatility through applications in pipeline robots, material handling, and dexterous hands.