Title: An efficient formulation of the Bayesian occupation filter for target tracking in dynamic environments

Authors: M.K. Tay, K. Mekhnacha, C. Chen, M. Yguel, C. Laugier

Addresses: INRIA Rhone-Alpes, Team e-Motion, Montbonnot, Saint Ismier Cedex, France. ' ProBayes, Inovallee, Saint Ismier cedex, France. ' Advanced Electronic Department (DEA), API : TCR RUC T 62, Technocentre Renault, 1 Avenue du Golf, Guyancourt cedex 78288, France. ' INRIA Rhone-Alpes, Team e-Motion, Montbonnot, Saint Ismier Cedex, France ' INRIA Rhone-Alpes, Team e-Motion, Montbonnot, Saint Ismier Cedex, France.

Abstract: The Bayesian Occupation Filter (BOF) has proven successful for target tracking in the context of automotive applications. This paper describes an improved BOF for target tracking with lower computational costs while retaining the key advantages of the original BOF formulation. The BOF takes the form of a grid-based decomposition of the environment. In contrast to the original BOF, each cell of the newly proposed BOF contains an additional distribution over the velocity of the propagating cell occupancy. This is estimated using Bayesian filtering. We propose how to deal with the grid discretisation problem. Object-based representations do not exist in the BOF grids. However, there are often applications which require the definition and tracking at the object level. A simple clustering and target tracking methodology is used to illustrate how to obtain this object level representation. Experiments based on tracking humans in indoor environment were conducted.

Keywords: Bayesian filtering; occupancy grids; target tracking; Bayesian occupation filter; grid discretisation; sensor fusion.

DOI: 10.1504/IJVAS.2008.016483

International Journal of Vehicle Autonomous Systems, 2008 Vol.6 No.1/2, pp.155 - 171

Published online: 31 Dec 2007 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article