Application of Artificial Intelligence Techniques to 6DOF Trajectory Analysis

摘要

The analysis of experimental free-flight trajectories consists in determining the causes of problems observed during the flight and providing recommendations to correct them. This task requires to match the observed motion to a predicted theoretical motion, which is very complex when using a six-degree-of-freedom (6DOF) model of the motion. Since only expert ballisticians are able to solve this task efficiently, the authors are developing an expert system which models the knowledge of a human expert. The main issues are to decompose the expert's high-level task into primitive tasks and to combine multiple levels of data analysis to improve the efficiency of the system.