Mission planning for joint suppression of enemy air defenses using a genetic algorithm

Abstract

In this paper we present a genetic algorithm applied to the problem of mission planning for Joint Suppression of Enemy Air Defenses (JSEAD) in support of air strike operations. The stochastic nature of JSEAD scenarios and the complexity of JSEAD operations and interactions make this an especially challenging problem within the military domain. JSEAD planners and analysts stand to benefit from any advances in tools that address this problem. While our interest in this subject is broad, in this paper we are specifically investigating methods for developing robust plans that include routes for JSEAD assets, target types, firing ranges, and take off time, subject to multiple objective functions that capture different aspects of mission performance. The multi-objective optimization is performed by the Dynamic Non-Dominated Sorting GA (DNSGA), a non-elitist variant of NSGA-II. The objective functions are evaluated using a stochastic agent-based JSEAD simulation, and we assess the quality of mission plans produced by the GA in a set of test scenarios. The results from these tests indicate that our approach has significant promise as a component of a JSEAD mission planning tool.