Discontinuous Trajectory Tracking of Hypersonic Glide Reentry Vehicle: An Intention Inference Approach

For the tracking of hypersonic gliding reentry vehicles (HGRVs), most existing methods are designed for continuous trajectories. However, in real-world scenarios, due to sensor limitations, discontinuous trajectories are commonly encountered, leading to the failure of tracking methods to converge promptly and causing substantial tracking errors. This problem results in a significant decline in the performance of current state estimation and prediction methods. In this article, a tracking algorithm based on intention inference is proposed for the scenarios of discontinuous trajectories. First, a multi-intention model incorporating attack intentions, no-fly zones, and ground threats is established to infer the sign of the bank angle. Second, the intention inference-based square-root cubature Kalman filter (Int-SRCKF) is designed by combining the bank angle inference results with the SRCKF. Then, for unobservable trajectory segments, the trajectory prediction based on the intention inference model is developed. Consequently, a state estimation and prediction framework for discontinuous HGRV trajectories is formed, which overcomes the limitations in real-world situations and improves the tracking performance under high-maneuverability conditions. Finally, the simulation results for both continuous and discontinuous scenarios are presented to demonstrate the effectiveness of the proposed tracking algorithm.