Guided bomb release planning based on Monte Carlo in a distributed virtual environment

Air-to-ground strike has become one of main forms of modem warfare, and the guided bomb release is an important and key part of the attack. The guided bomb release planning aims to accomplish a precise target hit in enemy's region while guarantee the pilot's safety. We propose a robust Monte Carlo method by taking error perturbations into consideration, comparing to the traditional sequential quadratic programming method under extreme conditions. At the same time using a distributed virtual physics environment we can obtain much more detailed realism relative to the conventional simulator running on a single machine. The experimental results verify that Monte Carlo methods can improve hit probability and weapon efficiency significantly. Furthermore, the 3D visualised environment plays a very important role in training pilots, so this simulator will decrease the cost and time requirements of physical experiment that are not always compatible with strict military task.