Performance Measures for Networks of Coupled Oscillators

Abstract

Networks of coupled oscillators provide a model for a diverse range of systems, such as power grids, vehicle platoons, and biological networks. Stability in these systems is generally related to consensus of the agents or coherence of the network. A related but less studied question relates to the performance of such systems. This work analyzes a series of nodal and network performance measures for systems of coupled oscillators, where the network performance is an aggregate measure across all oscillators in the network, and the nodal performance measure considers the performance of two nodes with respect to one another. We connect the nodal performance measure to the concept of network coherence in interconnected subsystems of vehicle platoons with local and global velocity feedback, and give analytical results for the special case of a subsystem with a tree structure. Additionally, we demonstrate how these concepts can be used to investigate transient losses in power grids.

Speaker Bio

Ted Grunberg received the B.S. degree in mechanical engineering from the Johns Hopkins University in 2013. He is currently pursuing the MSE degree in mechanical engineering at JHU under the supervision of Professor Dennice Gayme. His research is focused on the analysis and control of networked systems.

and

A State-based Approach to Safety of Component-based Medical Robot Systems

Abstract

A variety of medical robot systems have been developed both in academia and in industry, and commercial products are actively used in modern operating rooms.

However, systematic methods for safety of these systems have not yet received much attention, whereas the scale and complexity of modern medical robot systems have been continuously increasing to perform given tasks in faster, safer, and more efficient manners. This makes it even more challenging to design, build, and test large and complex medical robot systems. As an approach to these issues, this work presents a state-based semantics that can explicitly define, capture, and test the run-time status of component-based medical robot systems.

Based on the semantics, a software architecture is designed to facilitate the system development process by leveraging the characteristics of component-based systems.

To demonstrate the benefits of the proposed methods, experience with the refactoring of the software stack of a commercial orthopaedic surgery robot is presented as a case study.

Speaker Bio

Min Yang Jung is a Ph.D. candidate in Computer Science at The Johns Hopkins University. He received his B.S. in Electrical Engineering and M.S. in Biomedical Engineering from Seoul National University, South Korea. His research is focused on the safety of component-based medical and surgical robot systems and he has been working on a robot system for orthopaedic surgery as part of his PhD. He is one of the main contributors of the cisst/SAW framework.