Adaptive MCMC particle filter for tracking maneuvering target

When a target performs maneuvering movements, the target´s state variables, such as position and velocity, vary and are not restricted to a fixed dynamic model. New features of posterior distribution of the target state are encountered during the tracking process. In the MCMC based particle filter methods, the particles are moved towards the target posterior distribution to adapt to the new features formed during the tracking process. However, the traditional MCMC sampling needs a lot of iterations to converge to the target posterior distribution, which is very slow and not suitable for real-time tracking problem. In order to speed the convergence rate, a new method named adaptive MCMC based particle filter method, which is the combination of the adaptive Metropolis (AM) method and the importance sampling method, is proposed to tackle the real-time tracking problem.