Output details

Deterministic resampling: Unbiased sampling to avoid sample impoverishment in particle filters

This paper extends the state-of-the-art of resampling (which is a hot-topic in the field of particle filters) by proposing a new method called Deterministic Resampling to avoid sample impoverishment. It has attracted worldwide citations in one and half years since its publication. It improves the accuracy of state estimation for non-linear and non-gaussian dynamical systems. It is efficient in multiple dimensional cases and obtains better accuracy than the basic resampling and roughening methods. It will find application in problems such as multi-dimensional ballistic object tracking and mobile robot localisation. Verification: Dr. Yang Gao (nchygy@126.com), Chang’an University, China.