基于模型预测静态规划的自适应轨迹跟踪算法

Trajectory tracking is one of the key methods for autonomous aircraft motion control. Tracking Model Predictive Static Programming (T-MPSP) is an efficient non-linear model-based trajectory tracking algorithm. However, the algorithm may also result in unsatisfactory tracking performance in the cases when the aircraft is impaired, and when the utilized aircraft model deviates significantly from the nominal model. An adaptive trajectory tracking algorithm is proposed based on real-time parameter estimation. The algorithm realizes the real-time parameter estimation under the framework of Model Predictive Static Programming (MPSP), providing computationally efficient closed-form solution; and updates the model used in the T-MPSP in real-time, thus effectively ensuring a better guidance accuracy. The proposed algorithm is an extension of the MPSP algorithm, and therefore has a high overall computational efficiency, so it is amenable for online applications. Numerical simulation validates the effectiveness of the proposed algorithm.