Undergraduate Research Opportunities in Manufacturing Engineering

Sensor Networks and Information Systems
Christos Cassandras, Professor (joint with ECE)
http://vita.bu.edu/cgc/CODES
cgc@bu.edu

Discrete Event Systems have emerged with the rapid evolution of computing, communication, and sensor technologies. They include computer and communication networks, automated manufacturing systems, air traffic control systems, command-control systems, and distributed software systems. A significant portion of the “activity” in these systems, sometimes all of it, is governed by operational rules designed by humans; their dynamics are therefore characterized by asynchronous occurrences of discrete events, some controlled (like hitting a keyboard key, turning a piece of equipment “on”, or sending a message packet) and some not (like a spontaneous equipment failure or a packet loss). These features lend themselves to the term “discrete event system.” When combined with traditional systems (cars, planes, manufacturing processes), they give rise to what are called Hybrid Systems. Professor Cassandras directs the Control of Discrete Event Systems (CODES) Laboratory which operates within the Boston University Center for Information and Systems Engineering (CISE). Modeling, analyzing the behavior of Discrete
Event and Hybrid Systems, and ultimately controlling them are challenges that Professor Cassandras and his students are researching within CISE and the CODES Lab.

Dynamic Power Management in Wireless Sensor Networks
In wireless network environments, especially sensor networks, the limited battery power available to nodes is a key limitation to
network lifetime and performance. We develop dynamic power management methods that exploit the trade-off between energy and execution speed. By intelligently varying the voltage used by nodes in executing tasks such as transmitting, receiving, and data
processing, one can often dramatically reduce power consumption at a small cost of slowing down task execution when this is warranted. We are developing a new experimental test bed of over 100 Motes (wireless nodes with sensing capabilities) that will be deployed around the Manufacturing Engineering building to perform experiments involving power management, localization, and fusion of data collected by the Motes.

Intelligent Simulation and Rapid Learning of Discrete Event and
Hybrid Systems
We develop simulation models of manufacturing systems, computer networks, sensor networks, and transportation systems not only to study their behavior but also to learn how to control them so as to optimize their performance. In a manufacturing system, for example, intelligent simulation techniques can be used to determine optimal lot sizes which can provide tremendous cost savings. Or they can be used to choose scheduling policies which can optimally trade off
equipment utilization against lead times.

Cooperative Control of Autonomous Vehicles
We control small wirelessly communicating robots to carry out various cooperative “missions.” For more details and sample
pictures and movies, see http://frontera.bu.edu/CoopCtrl.html