Control of dynamic systems in spatial networks: Applications, results, and challenges

Abstract

A dynamic network is characterized by a spatially distributed set of nodes with sensors and actuators which measure and effect the dynamic system. These nodes are coordinated through the performance objectives and the possible dynamic coupling between the nodes in the presence of large uncertainties in the environment, in the observations, and through the communication channels between nodes. Examples of spatial networks that will be presented are: The control of a formation of aerial vehicles that reduces the total drag to approximately that of an equivalent slender wing constructed by joining together all the wings in the formation; the development of fault-tolerant control systems for the intelligent vehicle highway system; and robust boundary layer control to reduce friction drag in both transitional and turbulent flows in channels and wings is obtained by a network of shear stress sensors and actuators along the boundary. The control of dynamic networks can be viewed as essentially in its infancy and general procedures based on computational algorithms used for implementation is a long term goal. However, near term goals and challenges can be posed which represent important elements of the more complete problem.