Skip navigation

Systems of Systems

Systems of Systems

W. A. Crossley
My research activities involve identifying and/or developing formal, repeatable approaches for design in the Systems of Systems (SoS) context. A significant challenge of design within an SoS is determining the appropriate mix of independent systems to meet a set of needs or provide a set of services. Comprehensive methods do not yet exist for allocation of resources that have variable properties. Because the constituent systems are capable of independent operation, they may collaborate or compete; design methods must incorporate aspects of game theory. Evolution of an SoS over time will be more efficient if design methods include making decisions about when to retire existing assets, when to add new assets, and when to upgrade assets with improved technologies; extensions of dynamic programming are needed for this. Systems of Systems are (and will be) designed and operate under uncertainties. This requires new, or greatly expanded, approaches for design under uncertainty and reliability-based design. Some additional details about these efforts are available at Prof. Crossley's personal web page:
D. DeLaurentis
My research is focused upon the development of foundational methods and tools for addressing problems characterized as System-of-Systems. Pursuit of this focus includes establishment of an effective frame of reference, crafting of a common lexicon and establishing a unified framework for addressing SoS problem classes. This includes the study of various modeling & simulation techniques, including probabilistic robust design (including uncertainty modeling/management), agent-based modeling, network theory, object oriented simulations, and tools for capturing the interaction of requirements, concepts, and technologies. The present context for the research is the exploration of Next-generation Air Transportation systems, especially including the presence of revolutionary aerospace vehicles, new business models, and alternate policy constructs. Defense, healthcare and space exploration applications are also under study.
I. Hwang

Dr. Hwang's research interests are focused on developing modeling, simulation, and control methodologies for large-scale networked embedded systems with high performance and high reliability, in which individual subsystems are connected through communication networks (system-of-systems). These systems have inherently decentralized structures, i.e., the overall system consists of multiple, heterogeneous, autonomous subsystems called agents that are capable of making their own decisions, sensing, and communication. Applications include air and ground transportation systems, computer controlled systems (e.g. Unmanned Aerial Vehicles (UAVs) and satellites), and communication systems (e.g. sensor networks and power grids).

These complex systems can be modeled hierarchically; each agent is a dynamical system with embedded discrete control logic and thus it is difficult to accurately model this kind of systems as either continuous dynamical systems or discrete event systems. In order to accurately describe two different aspects of the behavior of the systems, Dr. Hwang is interested in the model of hybrid systems. Hybrid systems are dynamical systems that combine continuous dynamics modeled by differential (or difference) equations describing the physical behavior of the system and discrete dynamics modeled by finite automata representing decision making logic. The over all system can also be modeled as a hybrid system comprising of a set of sub-hybrid systems.

Dr. Hwang's research topics include modeling, analysis, simulation, and control of hybrid systems with applications to: (1) Air Traffic Control such as future National Airspace Systems design (mixed air space, free flight, etc.), autonomous conflict detection and resolution, pilot's intent inference and conformance monitoring, and airspace security monitoring and safety verification; (2) coordination of multiple-vehicle systems such as a group of UAVs and satellites; (3) sensor networks; and (4) health monitoring of large-scale structural systems.

D. Sun
My research areas include air traffic control, strategic traffic flow management, dynamic airspace configuration, and studies for Next Generation Air Transportation System (NextGen). I am interested in mathematical modeling and control of physical networked dynamical systems. Constraints are inherent to physical problems, yet they often make mathematical formulations hard to solve. I am interested in developing algorithms to solve these constrained optimization problems, in particular, algorithms to solve large scale problems that were beyond the reach of standard techniques. With an emphasis on aerospace engineering, I am interested in advancing the state of science in modeling and solution of stochastic, dynamic air traffic flow management problems, helping define and build the NextGen traffic flow management vision, philosophy, and architecture. I am also interested in implementing my algorithms on real-time platforms and building toolkits (software) for decision support. For air traffic control, the long term goal of the toolkit is the partial automation of some functions currently performed by humans, with the objective of improving safety and performance while easing workload.