Preliminary Results in Sliding Autonomy for Coordinated Teams

We are developing a coordinated team of robots to assemble structures. The assembly tasks are sufficiently complex that no single robot, or type of robot, can complete the assembly alone. Even with a group of multiple heterogeneous robots, each adding its unique set of capabilities to the system, the number of contingencies that must be addressed for a completely autonomous system is prohibitively large. Tele-operating a multiple robot system, at the other extreme, is difficult and performance may be highly dependent on the skill of the operator. We propose and evaluate an implementation of a framework that, ideally, provides the operator with a means to interact seamlessly with the autonomous control system. Using an architecture that incorporates sliding autonomy, the operator can augment autonomous control by providing input to help the system recover from unexpected errors and increase system efficiency. Our implementation is motivated by results from an extended series of experiments we are conducting with three robots that work together to dock both ends of a suspended beam.