Abstract:
In unstructured environments, ant colonies demonstrate remarkable abilities to adaptively form functional structures in response to various obstacles, such as stairs, gaps, and holes. Drawing inspiration from these creatures, robot swarms can collectively exhibit complex behaviors and achieve tasks that individual robots cannot accomplish. Existing modular robot platforms that employ dynamic coupling and decoupling methods often suffer from high energy consumption, limited mobility, and insufficiently robust coupling mechanisms for forming functional structures.

This thesis presents the Puzzlebot system, which takes inspiration from the adaptive abilities of ant colonies in unstructured environments. Robots can dynamically couple and decouple with each other with passive coupling mechanisms. We define passive coupling mechanisms as those that require no dedicated energy for coupling. With rigid and soft passive connections, we demonstrated that the Puzzlebots can dynamically form bridges across gaps and flexible chains that traverse stairs. To further study the functionality of coupled structures, we propose to design and iterate the coupling mechanisms and conduct experiments on robots collaborating to navigate on rough terrain, and pushing and organizing cluttered objects. For passive coupling mechanisms without sensors and actuation, it is essential to model the connection state externally for accurate control and planning. We propose to use a particle filter-based approach to infer the connection states between robots.

Thesis Committee Members:
Katia Sycara, Co-chair
Zeynep Temel, Co-chair
Aaron Johnson
Nikolaus Correll, University of Colorado at Boulder

More Information