Customizable bistable units for soft-rigid grippers enable handling of multi-feature objects via data-driven design

Area of Science:

• Robotics and Materials Science
• Development of novel soft-rigid grippers integrating bistable structures for advanced robotic applications.

Background:

• Soft-rigid grippers combine soft and rigid components for versatile object manipulation.
• Bistable structures offer rapid-response grasping and self-locking but face challenges with multi-feature objects.

Purpose of the Study:

• To develop customizable bistable units for soft-rigid grippers to handle irregular-shaped, fragile, and variable-weight objects.
• To create a data-driven design framework for optimizing bistable unit performance.

Main Methods:

• Utilizing customizable bistable units with tailored contact blocks (CBs) for grasping force control.
• Employing an inverse design framework with extremely randomized trees (ERT) and differential evolution (DE) algorithms for CB design.
• Developing an ERT model to accurately predict structural deformation, material behavior, and contact dynamics.

Main Results:

• Achieved 96.4% prediction accuracy in the ERT model for bistable unit behavior.
• Demonstrated grasping force control for objects with varying fragility and weight.
• Enabled overload protection and shape-conforming reconfiguration for irregular objects.

Conclusions:

• The novel bistable unit design enhances soft-rigid grippers' flexibility, scalability, and efficiency.
• This approach offers new possibilities for grasping complex objects in robotic technologies.