CFD based parameter tuning for motion control of robotic fish

Area of Science:

• Robotics
• Fluid Dynamics
• Biomimetics

Background:

• Robotic fish struggle with agile swimming due to complex fluid environments and limited battery life.
• Directly tuning motion control parameters on physical robotic fish is challenging.

Purpose of the Study:

• To develop and validate a computational fluid dynamics (CFD) simulation platform for optimizing robotic fish motion control parameters.
• To enhance the swimming ability and efficiency of robotic fish through simulation-guided parameter tuning.

Main Methods:

• A CFD simulation platform was created, incorporating a computational robotic fish with morphology and gait control inspired by real fish.
• Gait control was implemented using a central pattern generator (CPG), and the CFD model used a hydrodynamic-kinematics strong-coupling method.
• The platform was tested with active disturbance rejection control (ADRC) and analyzed power costs and swimming efficiency.

Main Results:

• Trajectory comparisons between simulated and real robotic fish confirmed the platform's effectiveness.
• The simulation platform allowed for robust tuning of motion control parameters.
• Analysis of power costs and swimming efficiency provided insights into optimized control strategies.

Conclusions:

• The developed CFD simulation platform is a powerful and robust tool for robotic fish design and motion control parameter optimization.
• This approach offers a novel method for improving the swimming performance and efficiency of robotic fish.
• Simulation-based tuning significantly overcomes the limitations of direct experimental parameter adjustment.