The improved unscented Kalman particle filter based on MCMC and consensus strategy

In the traditional Particle Filter algorithm, there is particle degradation and tracking accuracy is not good, so a new improved unscented particle filter algorithm with the Markov Chain Monte Carlo (MCMC) and consensus strategy is discussed. The algorithm uses unscented Kalman filter to generate a proposal distribution, which incorporates the latest observations into a prior updating routine. And the algorithm utilizes MCMC sampling method to make the particles more diversification. Meanwhile, the algorithm is optimized by consensus strategy, which makes the state estimates of all network nodes converge to a more precise value. The simulation results show that the improved unscented Kalman particle filter solves particle degradation effectively and improves tracking accuracy.