Solving Combat Control problem using Soft Computing approach

This paper deals with a generic solution for combat control problem which is generally addressed using conventional methods and human decision making. The aim of this paper is to develop the soft computing approach to automate the decision making process for combat control operations. The automatic combat control system design usually consists of four stages. The first stage is to select the sensor or sensors to produce the target measurements. The next stage is the preprocessing of the data and the location of regions of interest within the data (segmentation). The third design step is feature extraction and selection. The last step is the processing of the features for decision making (classification and combat). This paper deals with only combat control part which is the most challenging problem to be solved in real battle field scenario. Usually this part is being done using decisions taken by human operators. In order to automate this process, generally hard classifiers are being used for combat control operations. These hard classification approaches does not give good performance in noisy environment. A relatively new and emerging soft computing approach, which is a combination of Neural Network and Fuzzy classification techniques, is proposed for performance improvement. The algorithm developed in this paper is tested for a typical combat control problem which deals with selecting a particular weapon against a target in optimum way. The soft computing approach proposed here has demonstrated a relatively complex decision making in noisy environment, with a large number of input parameters.