Distributed implementations of particle filters

Particle filtering has a great potential for solving highly nonlinear and non-Gaussian estimation problems, generally intractable within a standard linear Kalman filtering based framework. However; the imple- mentation of particle filters (PFs) is rather computation- ally involved, which nowadaysprevents them frompractical real-world application. A natural idea to make PFs feasi- ble for "real-time" data processing is IO implement them on distributed multiprocessor computer systems. This paper presents three schemes for distributing the computations of generic particle filters, including resampling and. option- ally, a Metropolis-Hastings (MH) step. Simulation results based on a maneuvering target tracking scenario show that distributed implementations can provide a promising solu- tion 10 the steep computational burden incurred when using a large number of particles.