Multipotent Systems: Combining Planning, Self-Organization, and Reconfiguration in Modular Robot Ensembles.

Area of Science:

• Robotics
• Artificial Intelligence
• Computer Science

Background:

• Mobile multirobot systems are increasingly vital across various disciplines, including industrial automation and search and rescue.
• Current systems are either homogeneous (robust but specialized) or heterogeneous (flexible but prone to single-point failures).
• A gap exists in unifying the advantages of both homogeneous and heterogeneous mobile robot systems.

Purpose of the Study:

• To present a novel reference architecture for mobile robots that bridges the gap between homogeneous and heterogeneous systems.
• To introduce the concept of 'multipotent systems' that combine the strengths of both homogeneous and heterogeneous configurations.
• To enable homogeneous robots to dynamically acquire specialized, heterogeneous capabilities at runtime.

Main Methods:

• Development of a reference architecture defining the technological interplay for multipotency.
• Integration of reconfigurable and self-descriptive hardware (sensors and actuators) for runtime capability adaptation.
• Incorporation of autonomy mechanisms and algorithms to support multipotency in robot ensembles.

Main Results:

• Demonstration of 'multipotent systems' that merge redundancy benefits with specialized task execution.
• Successful creation of 'ensembles' where multiple multipotent robots cooperate.
• Validation of the approach through real-world experiments and simulation environments.

Conclusions:

• The proposed reference architecture successfully enables the creation of multipotent systems and ensembles.
• This approach allows homogeneous robots to achieve heterogeneous specialization dynamically, enhancing system versatility.
• The research provides a foundational framework for more adaptable and robust mobile multirobot applications.