User issues associated with sensor fusion and ATR
John M. Irvine · 2003
Growing military requirements and shorter timelines are placing greater demands on Intelligence, Surveillance, and Reconnaissance (ISR) assets. At the same time, advances in sensor technology have vastly increased the quantity and types of imagery data available to support ISR missions. Together, these factors are driving toward greater reliance on automated processing and exploitation techniques, such as automated target recognition (ATR) and sensor fusion. While there are substantial challenges associated with the development of effective methods for processing sensor data, a critical component is the role of the human operator. Often, ATR and sensor fusion technology is designed to aid the analyst, but the jinal decision rests with the human. The role of the human raises two related sets of issues: Outimization of the Interface: A system that accurately extracts information from sensor data is only of value if the information is successfilly conveyed to the user. The design of the human-computer interface, presentation and visualization of results, and the control that the user exerts over the jlow of information are all critical factors in the development of a successful system, Evaluation with the Human in the Loon: Assessing the benefits of ATWsensor jiisions assistance raises interesting methodological challenges. We will review the critical issues associated with evaluations of human- in-the-loop systems and present a methodology for conducting these evaluations. Experimental design issues addressed in this discussion include training, learning effects, and human factors issues.