Mechanical multiaxis force sensor for directly bridging sensing and fluidic actuation

Area of Science:

• Robotics
• Materials Science
• Fluid Mechanics

Background:

• Robots rely on sensors and actuators for operation, often using complex control loops.
• Reactive architectures offer robust and efficient sensorimotor control by directly linking sensing to actuation.
• Existing systems may require significant computational resources and external energy input.

Purpose of the Study:

• To introduce a novel fluidic-based multiaxis mechanical soft force sensor (ME-SOFS) as a mechanical analogy for reactive systems.
• To demonstrate a system that directly couples sensory signals with fluidic actuation, eliminating external computation.
• To enable soft robots with advanced somatosensory force sensing capabilities.

Main Methods:

• Development of a fluidic-based multiaxis mechanical soft force sensor (ME-SOFS).
• Utilizing fluid transduction to convert applied force into mechanical output for fluidic actuators.
• Integration of the ME-SOFS into robotic systems for sensing and actuation.

Main Results:

• The ME-SOFS directly couples sensory input with fluidic actuation, removing the need for external computation or energy.
• Demonstrated applications include directional droplet manipulation, controlled bending of cilia arrays, and enhanced robotic grasping learning via haptic feedback.
• The system successfully realized multiaxis force sensing for soft robots.

Conclusions:

• Fluidic approaches can enable simplified, closed sensing-actuation loops for soft robots.
• The ME-SOFS technology facilitates the development of robots with enhanced somatosensory capabilities.
• This work paves the way for advanced haptic human-machine interfaces and more intuitive robotic control.