Virtual Target Selection for a Multiple-Pursuer-Multiple-Evader Scenario

作     者：Weintraub, Isaac E. Von Moll, Alexander Casbeer, David W. Manyam, Satyanarayana G. 

作者机构：Air Force Res Lab Control Sci Ctr Wright Patterson AFB OH 45433 USA InfoSciTex Corp 4027 Col Glenn Hwy Dayton OH 45431 USA 

出 版 物：《JOURNAL OF AEROSPACE INFORMATION SYSTEMS》 (J. Aerosp. Inf. Sys.)

年 卷 期：2025年第22卷第5期

页      面：391-400页

核心收录：

基　　金：The authors would like to thank Adam Gerlach for providing assistance with using the Julia programming language. DISTRIBUTION STATEMENT A. Approved for public release: distribution unlimited (AFRL-2024-5685). Cleared 09 Oct 2024 

主　　题：Mixed Integer Linear Programming Optimal Control Theory Optimization Algorithm Weapon Target Assignment Monte Carlo Analysis Guidance, Navigation, and Control Systems Missile Defense 

摘      要：This paper considers an N-pursuer-M-evader scenario involving L virtual targets. The virtual targets serve as an intermediary target for the pursuers, allowing the pursuers to delay their final assignment to the evaders. However, upon reaching the virtual target, the pursuers must decide which evader to capture. It is assumed that there are more pursuers than evaders and that the pursuers are faster than the evaders. The objective is two-part: first, assign each pursuer to a virtual target and evader such that the pursuer team s energy is minimized, and, second, choose the virtual targets locations for this minimization problem. The approach taken is to consider the Apollonius geometry between each pursuer s virtual target location and each evader. Using the constructed Apollonius circles, the pursuer s travel distance and maneuver at a virtual target are obtained. These metrics serve as a gauge for the total energy required to capture a particular evader and are used to solve the joint virtual target selection and pursuer-evader assignment problem. This paper provides a mathematical definition of this problem, the solution approach taken, and an example. Following the example, a Monte Carlo analysis is performed, demonstrating the efficacy of the algorithm and its suitability for real-time applications.