AI-based Air-to-Surface Mission Planning using Predictive Launch Acceptability Region Approach

In this paper, a dynamic air-to-surface mission planning strategy based on artificial intelligence (AI) is proposed for targets of opportunity in order to guide the pilot to follow the shortest and safest trajectory by taking advantage of recent advances in military technologies like predictive launch acceptability region (LAR) approach and High Speed 1760. All of the surface targets of an air-to-surface mission are usually planned and loaded to aircraft before the mission. However, sometimes pilots may be suggested to destroy some unanticipated targets which are unplanned. In that case, pilots can be obliged to deviate from the waypoints of the planned mission in order to accomplish the new unplanned task and this can pose a great danger since there could be many potential threats around theater of war. In proposed method, surface threats and predictive launch acceptability region queries are modeled. Then, probabilistic roadmap algorithm with Dubins distance is applied to produce a waypoint trajectory which makes possible to reach closest goal state. The proposed method is proven by constituting a realistic simulation environment based on ROSplane considering mechanical and environmental factors. In total 25 flight simulations, the maximum observed deviation and the mean deviation from a waypoint are calculated as 11.1 meters and 2.2 meters respectively in a 5000m x 5000m x 300m space. Therefore, the results show that the proposed method can dynamically generate waypoint trajectories by using predictive launch acceptability region queries, which are safe and possible to follow.