An adaptive multivariate Student's t‐process recursive method for hypersonic glide vehicle trajectory prediction

The trajectory prediction of the hypersonic glide vehicle (HGV) can provide hit point information for early warning systems, which is great significance near-space defence operations. However, heavy-tailed noise caused by abnormal environmental disturbance and abrupt changes in seriously affects accuracy HGV prediction. To solve problem under noise, we propose an adaptive multivariate Student's t-process regression method called aEM-MVTPR. Firstly, a model developed using t-distribution. Secondly, (MVTPR) derived, learn time series features mine correlations among variables. Finally, to further improve method's robustness, use accelerated expectation maximisation algorithm Pearson correlation analysis adaptively estimate adjust initial values MVTPR parameters. simulation experimental results show that proposed has more accurate predicted variables than Gaussian process (MVGPR) conditions. In addition, ability aEM-MVTPR parameter makes it robust different environments.