This will be a hybrid event with in-person attendance in Levine 307 and virtual attendance on Zoom. This week’s speaker will be virtual.
Various applications for autonomous multi-robot systems, such as disaster response, infrastructure repair, agricultural operations, contaminant cleanup, and wildfire suppression, will require these systems to act as distributed sensors, sources, and manipulators in uncertain environments, often without global localization or extensive inter-robot communication. This talk will present dynamical model-based approaches to controlling mobile robot collectives and soft modular robots that operate under such constraints using local sensing and local or no communication. By applying feedback control and optimization techniques to models that describe the dynamics of a continuous abstraction of the system and its individually controllable components, including their stochastic behaviors, scalable control strategies can be designed for a variety of objectives. This talk will discuss strategies for the particular objectives of mapping and control of scalar environmental fields by robotic swarms, cooperative object transport by groups of mobile robots, and reconfiguration of multi-segment continuum robots. The control strategies are illustrated in numerical and physics-based simulations and in experiments with robotic hardware.