Simulation-Based Air Mission Evaluation with Bayesian Threat Assessment for Opposing Forces

Abstract

Several advancements have been made to the air mission planning process in recent years, spawning several software tools that allow for a quick analysis of the feasibility of the mission. However, the determination of the most likely outcomes of an air mission plan is still a challenge for modelers and planners. Mathematical models are capable of representing the complexity of sensors and weapons systems, but are not as effective in providing a thorough visualization of the mission, as well as in accounting for environmental factors and interactions between multiple systems within an operational setting.

Our research addresses the Systems Engineering problem of enhancing a system’s response by predicting the adversarial reactions to different inputs. We adopt a physics-grounded simulation approach that focuses on the handling of uncertainty in behavioral models as a means of properly assessing the responses. As part of this research, the work presented in this paper proposes the use of high-resolution simulation for evaluating the survivability and mission accomplishment rates of an attacking aircraft during an incursion. The work differs from purely scripted simulation, which employs rule-based entities, predefined routes and behaviors for blue and red forces. Instead, our prototype relies on a Bayesian threat assessment and response methodology. Our goal is to represent the enemy’s decision process, while being able to predict the mission outcomes and to correctly assess the regions of higher vulnerability.