In order to solve the problem of poor accuracy and real-time performance of the traditional numerical integration method in solving the projectile landing position, the XGBoost optimized by the improved Dung beetle optimizer (IDBO) is proposed to improve the accuracy and efficiency of the projectile landing point information prediction. Firstly, the six-degree-of-freedom ballistic equation is established, and the fourth-order Runge-Kutta method is employed to solve it. Subsequently, the flight characteristics of the projectile are analyzed and the dataset is constructed. Then, the data is preprocessed and the IDBO-XGBoost projectile drop point model is established. Meanwhile, XGBoost projectile impact point prediction models optimized by Sparrow algorithm (SSA), Particle swarm optimization (PSO), Dung beetle optimizer (DBO) and IDBO were compared. The simulation results show that the prediction error of IDBO-XGBoost model is significantly lower than that of other models, and the accuracy is improved by an order of magnitude, which provides support for the prediction of projectile impact point and the research of remote fire precision strike.