A New Improved Particle Filter Algorithm Based on UKF and GASA

The degeneracy is the critical problem existed in particle filter (PF). In order to solve this problem, we propose a new algorithm combined PF with unscented Kalman filter algorithm (UKF) and genetic simulated annealing algorithm (GASA) in this paper. In the new algorithm, UKF is used to generate the importance proposal distribution which can match the true posterior distribution more closely, and GASA based on the survival-of-the-fitness principle is applied to enhance the diversity of samples. As a result, the simulation results indicate that the new algorithm can resolve the problem of sample degeneracy successfully and outperform other particle filter algorithms in terms of accuracy and suppression the noise.