This short investigation focused primarily on mechanisms that allow the robot to move about cheaply and effectively. My undergraduates and I rapidly prototyped many versions of robots that could provide two elements of functionality: mobility in the ceiling, allowing it to move across the entire building while remaining connected to power and the power transfer mechanism, allowing the power to be distributed across the grid in a reasonable way. The robots were connected to low-power leads that allowed it to operate effectively for long periods of time.
Our prototypes focused on two types of systems: rail-based and magnet based. The rail-based method allowed us to design more complex systems as they were explicit modifications of the environment itself while the non-invasive magnet based one was sensitive to major gradation changes (steps). One of our ideas that we never attempted was to get it to work with a ferromagnetic paint. While we ran out of time, I’m still very passionate about getting these types of robots working effectively in our environments.