Improved GA-based ICF target pose measurement

In Inertia Confinement Fusion (ICF) physical experiments, the accuracy of target positioning affects the successful rate of target hitting directly. A 3-CCD camera system is often used for tiny target measurement in ICF target positioning. Most of the current pose measurement methods utilize the well-known digital image processing technology to extract the target features in each image, then calculates the target's spatial coordinate and rotation matrix by integrating the feature values from three CCDs. Therefore, feature extraction errors in each image are superimposed in final result, which reduces the pose measurement precision. In this paper, we propose a solid model-based method which matching the target as a whole by the grey values in each image without utilizing image processing technology. The solid model matching optimistic problem is solved by an improved genetic algorithm (GA), called adaptive GA. Experiment is performed by using a 3-CCD camera system with general GA and adaptive GA respectively, the result shows the effectiveness of our adaptive GA in improving speed and accuracy.