Modeling, Analysis, and Computational Design of Muscle-driven Soft Robots.

Area of Science:

• Robotics
• Continuum Mechanics
• Computational Design

Background:

• Muscle-driven soft robots (MDSRs) are crucial for animal motion and soft robotics, but general kinematic modeling and design methods are lacking.
• Existing soft robot development has not adequately addressed the specific challenges of MDSRs.

Purpose of the Study:

• To present a comprehensive framework for kinematic modeling and computational design of homogeneous muscle-driven soft robots (MDSRs).
• To address the inadequacy of current design methods for MDSRs by providing a systematic approach.

Main Methods:

• Utilized continuum mechanics theory, including deformation gradient tensor and energy density function, to describe soft body characteristics.
• Employed a triangular meshing tool and piecewise linear hypothesis for discretized deformation analysis.
• Developed deformation models for MDSRs based on hyperelastic material constitutive modeling and addressed computational design through kinematic models and algorithms for parameter inference and muscle optimization.

Main Results:

• Established effective deformation models for MDSRs considering both internal muscle units and external driving points.
• Proposed algorithms for inferring design parameters from target deformations and optimizing muscle selection.
• Successfully developed and experimentally validated several MDSRs, demonstrating the framework's effectiveness through quantitative comparison of computational and experimental results.

Conclusions:

• The presented framework for deformation modeling and computational design of MDSRs is effective for creating soft robots with complex deformations.
• This approach can significantly facilitate the design of advanced soft robots, including those with applications like humanoid faces.