Fibrous contextual embodiments formed by task-invariant continuous blow spinning in dynamic environments.

Area of Science:

• Robotics
• Materials Science
• Morphological Computation

Background:

• Navigating dynamic environments poses significant challenges for robots.
• High-level robotic control is limited by environmental complexity, necessitating morphological adaptation.

Purpose of the Study:

• To develop a method for robots to create adaptable physical structures in situ.
• To enable robots to leverage environmental dynamics for morphological computation.

Main Methods:

• Introduced a model-free melt-blown process for creating unsupported thermoplastic webs.
• Demonstrated in-situ fiber spinning on moving supports, yielding highly elastic and stretchable webs.
• Explored task-invariant fiber spinning for various applications.

Main Results:

• Achieved webs with ~60% reversible strain and ~1,000% stretch at rupture.
• Successfully demonstrated applications including pathway creation, object immobilization, material sampling, and resource capture.
• Showcased the ability of spun fibrous bodies to emerge from and interact with environmental dynamics.

Conclusions:

• In-situ spun fibrous bodies enable robots to dynamically extend, adjust, and repair their physical architecture.
• This approach allows robots to morph context-specific components and tackle open-ended tasks by integrating environmental information.
• Robotic morphology can be dynamically adapted using environmental interactions, inspired by spiderweb structures.