|
Abstract : |
We present a statistical approach to adapting the sample set size of particle filters on-thefly. The key idea of the KLD-sampling method is to bound the error introduced by the samplebased representation of the particle filter. Thereby, our approach chooses a small number of samples if the density is focused on a small subspace of the state space, and it chooses a large number of samples if the state uncertainty is high. Both the implementation and computation overhead of this approach are small. Extensive experiments using mobile robot localization as a test application show that our approach yields drastic improvements over particle filters with fixed sample set sizes and over a previously introduced adaptation technique. 1, |
