Graphics Processing Unit Implementation of the Particle Filter

Modern graphics cards for computers, and especially their graphics processing units (gpus), are designed for fast rendering of graphics. In order to achieve this gpus have a parallel architecture which can be exploited for general-purpose computing on graphics processing units (gpgpu) as a complement to the central processing unit (cpu). In this paper gpgpu techniques are used to implement state-of-the-art recursive Bayesian estimation using particle filters (pf). The main steps of the pf are highly parallel except for the resampling step, and this paper discuss one way to handle this. The performance of this implementation is compared to that achieved with a traditional cpu implementation. The resulting gpu filter is faster with the same performance as the cpu filter for many particles.