Imperial College London > Talks@ee.imperial > CAS Talks > Parallel resampling for Particle Filters on FPGAs

Parallel resampling for Particle Filters on FPGAs

Add to your list(s) Download to your calendar using vCal

If you have a question about this talk, please contact Grigorios Mingas.

Particle Filters, also known as Sequential Monte Carlo (SMC) methods, are a set of simulation-based approach to compute posterior distributions and perform inference of unknown quantities from observations. Because of its outstanding performance and simplicity in implementation for nonlinear and/or non-Gaussian dynamic systems, it has been widely used in application fields such as target tracking, digital signal extraction, air traffic management and robot localisation. Resampling prevents the filter from weight degeneracy and improves the estimation of states by concentrating particles into domains of higher posterior probability. However, resampling is computationally intensive when the number of samples is large and, most importantly, it is not obviously parallelizable like the other steps of the particle filter. This talk gives a quick introduction on how to parallel resample on hardware especially FPG As, and presents our recent work on the parallel implementations of the state-of-the-art resampling algorithms in FPGA . The speedup over GPU designs will be also demonstrated.

This talk is part of the CAS Talks series.

Tell a friend about this talk:

This talk is included in these lists:

Note that ex-directory lists are not shown.

 

Changes to Talks@imperial | Privacy and Publicity