Dr. C. Allen Butler
Dr. Butler holds a B.A. in Mathematics from Texas Tech University and a PhD in Mathematics from the University of Illinois, Champaign-Urbana. He has been employed at Daniel H. Wagner Associates, Inc. since 1987 and is currently the President of the company. Throughout his career with Wagner Associates, Dr. Butler has served as the principal investigator for Department of Defense projects involving a variety of mathematical disciplines as applied to areas such as tracking, track correlation, data fusion, and search optimization. He has been involved in the development and implementation of optimal search techniques for a number of projects, most recently a research effort whose goal is the interdiction of narcotics smugglers in the Caribbean. Dr. Butler is an internationally recognized expert in Data Fusion. He served on the Navy’s Multi-Sensor Integration System Engineering Team (MSI SET), where he helped to perform a technical assessment of the current state-of-the-art in Multi-Sensor Integration for use in the next generation Shipboard Combat System. He also served as an expert Data Fusion consultant to the congressionally mandated “MTI-IMINT Fusion Study” performed by the Joint C4ISR Decision Support Center. Dr. Butler is the chair of the INFORMS Prize Committee for the “Daniel H. Wagner Prize for Excellence in Operations Research Practice” and currently serves as the secretary/treasurer for the Business Industry Government Special Interest Group of the Mathematical Association of America (BIG- SIGMAA).
The Mathematics of Data Fusion
In recent years, tremendous strides have been made in the improvement of existing and the development of new, more powerful, sensor systems. The result is a tidal wave of data which threatens to overwhelm the user, rather than assist her. The process of automatically filtering, aggregating and extracting the desired information from multiple sensors and sources is an emerging technology, commonly referred to as Data Fusion. In this talk, I will show how a wide variety of mathematical techniques are applied in this new discipline. I will begin with a discussion of the state estimation problem – determining the current position and velocity of an object based on a set of discrete observations (e.g. radar tracking of an aircraft). I will discuss a number of filtering techniques, including the a-b filter, the classic Kalman Filter, the Extended Kalman Filter and the Unscented Kalman Filter. I will then discuss the data association problem – given a set of observations or measurements taken over a period of time, determine which ones originate from the same real-world object. Finally, I will conclude with a discussion of Data Fusion Measures of Performance, attempting to answer the question, “How do you grade a system that produces probability distributions for answers?”