|Title||Building adaptive computer generated forces: The effect of increasing task reactivity on human and machine control abilities|
|Publication Type||Conference Paper|
|Year of Publication||2001|
|Authors||Bugajska, MD, Schultz, AC, Trafton, JG, Mintz, FE, Gittens, S|
|Conference Name||2001 Genetic and Evolutionary Computation Conference - Late Breaking Papers|
|Conference Location||San Francisco, CA|
Computer Generated Forces (CGF), in order to be effective training tools, must exhibit robust, challenging, as well as realistic behaviors. CGF tasks usually have both cognitive and reactive aspects to them. The reactivity has to co-exist with the "higher-level" cognitive activities like planning and strategy assessment, in the system that interacts with the environment. The overall purpose of our research is to merge a machinelearning algorithm (SAMUEL, an evolutionary algorithm-based rule learning system) with a cognitive model (ACT-R) into a system where the learning algorithm handles the reactive aspects of the task and provides an adaptation mechanism, and where the behavior’s realism is constrained by the cognitive model. Such a system would learn through experience so that it can adapt to changes in adversaries’ strategies and capabilities, to present human opponents with more exciting, varied, yet realistic training situations. This preliminary work presents an initial examination of effect of the changes in task reactivity on the human and SAMUEL control abilities.
NRL Publication Release Number: