The Coordinate Particle Filter - A novel Particle Filter for High Dimensional Systems

TL;DR

The paper introduces the Coordinate Particle Filter, a new approach that improves computational efficiency in high-dimensional systems by updating particle weights dimension by dimension, especially when not all dimensions are highly correlated.

Contribution

It presents the Coordinate Particle Filter, a novel method that reduces computational complexity by recursive, dimension-wise weight updates, enabling better performance in high-dimensional filtering tasks.

Findings

Significant performance improvement over traditional Particle Filter in high-dimensional scenarios.

Effective in systems where not all dimensions are highly correlated.

Validated on simulated and real data for multi-object and robotic manipulator tracking.

Abstract

Parametric filters, such as the Extended Kalman Filter and the Unscented Kalman Filter, typically scale well with the dimensionality of the problem, but they are known to fail if the posterior state distribution cannot be closely approximated by a density of the assumed parametric form. For nonparametric filters, such as the Particle Filter, the converse holds. Such methods are able to approximate any posterior, but the computational requirements scale exponentially with the number of dimensions of the state space. In this paper, we present the Coordinate Particle Filter which alleviates this problem. We propose to compute the particle weights recursively, dimension by dimension. This allows us to explore one dimension at a time, and resample after each dimension if necessary. Experimental results on simulated as well as real data confirm that the proposed method has a substantial performance advantage over the Particle Filter in high-dimensional systems where not all dimensions are highly correlated. We demonstrate the benefits of the proposed method for the problem of multi-object and robotic manipulator tracking.