Improved diagonal interacting multiple model algorithm for manoeuvering target tracking based on H∞ filter

This study is devoted to the problem of state estimate of discrete-time stochastic systems with Markov jump parameters. A robust algorithm - diagonal interacting multiple model algorithm based on H_∞ filtering is presented for manoeuvering target tracking when noise of measurement is of unknown statistics. Extensive Monte Carlo simulations show the effectiveness and superiority of the proposed algorithm.