The interconnection of sensors, actuators, and controllers enables local and remote supervision of physical environments. A fundamental difference between the traditional and the networked control problem is that the network introduces unreliable and time-dependent levels of service (delays, jitter, losses). In turn, network vagaries can jeopardize the stability, safety, and performance of the controlled units. The primary objective of this research is to devise integrated control and communication algorithms to compensate for the vagaries of network service.
Such control and communication strategies are targeted toward the application-layer and their objective is to deal with packet losses, delays, jitter, and network unreliability. This objective necessitates an integrated approach that combines networking (e.g., network measurement, modeling, and simulation; reliability) with decisions based on sensor data (e.g., feedback control). Although strategies have to be adapted to the specific application areas, the research addresses a general methodology by drawing from the foundations of Systems Theory and of Networking. Control algorithms and communication protocols are evaluated according to a methodology that combines analytical characterization, co-simulation, and prototype construction and validation.
The broader impact of this research includes teaching, course and curriculum development, graduate advising, and undergraduate research experiences. An objective of this work is to expand the presence of groups that are under-represented in Engineering and Computer Science. The investigators have on-going relationships with industry. The results of this research are published and freely disseminated.