Intent based trajectory prediction by multiple model prediction and smoothing

Trajectory prediction is important for trajectory-based operations in the next generation air transportation system. The intent of an en-route aircraft can benefit significantly the performance of trajectory prediction. An intent based trajectory prediction (IBTP) algorithm for civilian aircraft is proposed in this paper. The predictor based on the Interacting Multiple Model (IMM) algorithm is guided by the spatial and temporal information in the intent. First, the trajectory is predicted based on IMM prediction, where the dynamic model with a heading angle closer to the intended direction is assigned a larger weight. Second, a pseudo-measurement based on the spatial information (e.g., position and speed) in the intended waypoint is generated to update/smooth the predicted trajectory. An illustrated example is provided and results are evaluated by Monte Carlo simulation.