Robust whole-hand spatial manipulation via energy maps with caging, rolling, and sliding

Area of Science:

• Robotics
• Mechanical Engineering
• Artificial Intelligence

Background:

• Traditional robotic hands primarily use finger tips, limiting dexterity.
• Human hand manipulation involves all inner surfaces for complex tasks.

Purpose of the Study:

• To explore whole-hand usage in robotic dexterous within-hand manipulation.
• To develop a novel robotic hand and control model for enhanced spatial manipulation.

Main Methods:

• Introduction of the Model B, a four-fingered robotic hand.
• Implementation of a potential energy-based motion model for hand-object system control.
• Application of a 3D energy model for robotic hand manipulability comparison.

Main Results:

• The Model B hand successfully demonstrated complex spatial caging manipulation.
• The hand utilized all inner surfaces, including the palm, for object control.
• Validation through extensive benchtop and robotic arm experiments with diverse objects.

Conclusions:

• The developed energy model and Model B hand enable advanced robotic manipulation.
• Whole-hand manipulation significantly increases robotic dexterity and task capability.
• This research advances robotic hands for complex, real-world applications.