Maneuvering Target Tracking Performance Online Monitoring Based on Fréchet Distance

In the application of maneuvering target tracking, online performance monitoring helps to optimize the way of radar resource allocation and improve tracking accuracy. To solve the problem of very high computation cost in the existing online monitoring methods based on information geometry, we propose an online monitoring method for tracking performance based on the Frèchet distance in this paper. The Frèchet distance under the multivariate Gaussian distribution takes into account the mean and covariance in the filtering, which can get a more reasonable error measurement. The advantage of the method is high computational efficiency, which is conducive to online evaluation. Simulation results show that the performance monitoring results of this method for maneuvering targets are equivalent to the shooting algorithm based on information geometry, but the computation cost is significantly lower than that of the shooting algorithm.