Decision Theoretical Approach to Pilot Simulation
Kai Virtanen, Tuomas Raivio, Raimo P. Hämäläinen
Abstract
Kai Virtanen, Tuomas Raivio, Raimo P. Hämäläinen
Abstract
We simulate and analyze pilot decision making in one-on-one air combat using an influence diagram. Unlike most of the existing approaches, an influence diagram graphically describes the factors of a decision process and explicitly handles the decision maker’s preferences under conditions of uncertainty. In the pilot decision model, the possible combat situations related to each maneuver alternative are associated with a probability and a utility. Influence diagram analysis produces a probability distribution of the overall utility that represents the successfulness of a maneuver and gives information to make rational maneuvering decisions. Sensitivity analysis determines the impacts of different factors on the outcome of the maneuvering decision. The effects of sensor information that will reduce the uncertainty of the model are evaluated using Bayesian reasoning. The model can be utilized in the analysis of a single decision situation or as an automated decision making system that selects combat maneuvers in air combat simulators. I.
OpenAlex reports 40 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
We simulate and analyze pilot decision making in one-on-one air combat using an influence diagram. Unlike most of the existing approaches, an influence diagram graphically describes the factors of a decision process and explicitly handles the decision maker’s preferences under conditions of uncertainty. In the pilot decision model, the possible combat situations related to each maneuver alternative are associated with a probability and a utility. Influence diagram analysis produces a probability distribution of the overall utility that represents the successfulness of a maneuver and gives information to make rational maneuvering decisions. Sensitivity analysis determines the impacts of different factors on the outcome of the maneuvering decision. The effects of sensor information that will reduce the uncertainty of the model are evaluated using Bayesian reasoning. The model can be utilized in the analysis of a single decision situation or as an automated decision making system that selects combat maneuvers in air combat simulators. I.
Key concepts: Computer science, Aerospace engineering, Simulation, Aeronautics, Engineering