Hybrid Offline-Online Configuration Planning Approach for Continuum Robots Based on Real-Time Shape Estimation

Area of Science:

• Robotics
• Control Systems
• Mechanical Engineering

Background:

• Continuum robots offer high flexibility for confined spaces.
• Their inherent nonlinearities and uncertainties pose motion planning challenges.

Purpose of the Study:

• To develop a hybrid offline-online framework for continuum robot motion planning.
• To address collision avoidance and structural constraints in complex environments.

Main Methods:

• Formulated planning as nonlinear optimization with collision and structural constraints.
• Used a co-evolutionary strategy with differential evolution (DE) for optimization decomposition.
• Implemented an unscented Kalman filter (UKF) for real-time shape estimation and safe distance monitoring.
• Designed an online configuration refiner for local adjustments based on shape feedback.

Main Results:

• Demonstrated the effectiveness of the hybrid planning framework.
• Achieved enhanced motion smoothness and safe motion performance in real-world tests.
• Successfully steered continuum manipulators through constrained spaces.

Conclusions:

• The proposed hybrid framework effectively manages nonlinearities in continuum robots.
• This approach enhances safe and smooth navigation in challenging environments.
• The method provides a robust solution for continuum robot motion planning.