Battlefield Target Grouping by a Hybridization of an Improved Whale Optimization Algorithm and Affinity Propagation

Abstract

Target grouping, which is essentially a data clustering problem, is a research hotspot in the field of battlefield situation assessment. To address an unknown battlefield environment, affinity propagation based on an improved whale optimization algorithm (APBWOA) is proposed from the perspective of clustering. First, we propose a whale optimization algorithm based on a chaotic map and nonlinear inertia weight improvement called CPIW-WOA, which uses an improved circle map to generate initial populations and introduces nonlinear inertia weights to improve its convergence efficiency. The test results on nine benchmark functions show that the CPIW-WOA algorithm has superior performance to existing methods. Second, based on the fact that it fully considers the weights of the attributes in a given sample, the weighted Mahalanobis distance is adopted to replace the Euclidean distance. In addition, the silhouette index is introduced to determine the optimal number of clusters. By iteratively updating through CPIW-WOA to search for the optimal settings, the limitation of manually entering specified parameters can be overcome. Test results on real-world datasets show that the new method is more accurate and effective than other methods; therefore, it can provide effective solutions with respect to battlefield target grouping.