Texas A&M Engineering researchers working to secure ports and waterways
COLLEGE STATION, Texas -- Increased security in American ports and waterways is the goal of two engineering professors in Texas A&M University's Department of Industrial and Systems Engineering. Dr. Wilbert E. Wilhelm, Mike and Sugar Barnes Professor, and assistant professor Dr. Yu Ding have received a $331,111 grant from the National Science Foundation (NSF) for their project, "Strategic design and tactical operation of surveillance sensor systems for ports and waterway security." The researchers aim to design surveillance sensor systems to assure robust security in ports and waterways, especially to deal with unauthorized smaller boats that can easily gain access to sensitive targets. The researchers are using the Houston Ship Channel as a trial case. "Many people do not realize how vulnerable our ports are to a variety of threats," Wilhelm said, "and how devastating an attack could be to a surrounding city -- for example, as poisonous gas is released from a damaged storage container." In the first part of the project, strategic design, Wilhelm is determining where to locate sensors to provide adequate surveillance. Types of sensors include television cameras; infrared cameras; radar; and sound, motion or heat detectors, and Wilhelm said he is designing a heterogeneous sensor system so as not to rely on a single type of sensor. "We're looking at interdiction," Wilhelm said, "so we have to look at various types of threats. For instance, a large ship or tanker has certain characteristics, such as traveling at a certain speed or maneuvering in certain ways. Each type of threat travels at a different speed." Wilhelm said the researchers have to understand which points to observe in order to interdict these threats. Then, the information is fed into a sensor location model to optimize the cost of buying, installing and maintaining the sensors. "The challenge is to develop algorithms to solve problems as large as what might be found in actual applications," Wilhelm said. The researchers also have to deal with uncertainty so that, in case a sensor fails, a threat can still be observed. That's known as fault tolerance capability and is the tactical part of the project. "Fault tolerance analysis," Ding said, "is, `What if there is a failure due to harsh environmental conditions or intentional tampering? What kinds of redundancy are needed and at what cost level?'" Ding said that with different sensors come different types of information and different uncertainties. By integrating the information in an intelligent way, a heterogeneous surveillance system could potentially be able to detect anomalies, distinguishing between "normal" and abnormal operation and behavior of a ship in open water as well as in a ship channel or waterway. "Each type of ship has a set of unique characteristics," Wilhelm said. "In a ship channel, the way a ship turns and maneuvers and how it has to slow down to turn are unique to each type of ship. So if a ship starts to turn or speed up unexpectedly, then maybe it's been commandeered and is being piloted awkwardly, and we need to detect that anomaly. "Additionally, ships of certain sizes go to certain locations in a port or waterway, so if we find a tanker upstream from where it's supposed to be, that's anomalous." "But," Ding said, "you also have to reduce the false-positive rate. With too many false positives, the sensor network becomes a nuisance rather than a safeguard, so how can we keep the probability of a false positive in check?" Wilhelm and Ding said that the mathematical models they are developing will give better understating of how a surveillance sensor network should work and allow better decision making by port and waterway security personnel. "We're using the Houston Ship Channel as a trial case," Wilhelm said, "but our models are generic so they can apply to each and every U.S. port and waterway." Wilhelm and Ding are also researchers in the Industrial and Systems Engineering Division of the Texas Engineering Experiment Station (TEES), the engineering research agency of Texas and a member of The Texas A&M University System.