Research on control strategy of pneumatic soft bionic robot based on improved CPG

Area of Science:

• Robotics
• Control Systems
• Biomimetics

Background:

• Soft robots offer unique advantages in maneuverability and adaptability.
• Effective motion control is crucial for the practical application of pneumatic soft robots.
• Existing control strategies often struggle with accuracy and anti-interference in complex environments.

Purpose of the Study:

• To propose an improved Central Pattern Generator (CPG)-based motion control strategy for pneumatic soft bionic robots.
• To enhance the accuracy and anti-interference capabilities of soft robot motion control.
• To develop a rhythmic motion control method applicable to pneumatic soft robots.

Main Methods:

• Constructed a two-layer neural network topology model using 22 coupled Hopfield neuron nonlinear oscillators.
• Developed a CPG-ANFIS-PID (Central Pattern Generator-Adaptive Neuro-Fuzzy Inference System-Proportional-Integral-Derivative) motion control strategy.
• Utilized offline learning and training of membership functions based on ANFIS for control strategy construction.

Main Results:

• The CPG-ANFIS-PID strategy demonstrated rapid response to input parameter changes across different swimming modes.
• The control strategy stably outputted smooth, continuous dynamic position signals.
• Experimental analysis of the robot prototype confirmed stable coupled waveform signal output and biological rhythm motion propulsion.

Conclusions:

• The proposed CPG-ANFIS-PID control strategy provides significant accuracy and anti-interference characteristics for pneumatic soft bionic robots.
• The strategy enables robots to exhibit maneuverability, flexibility, and environmental adaptability.
• This work establishes a viable control strategy and rhythmic motion control method for pneumatic soft bionic robots.