New likelihood updating for the IMM approach application to outdoor vehicles localization

This paper presents the problematic of outdoor vehicle localization under the IMM (interacting multiple model) approach. The IMM is now a well known modular approach, which is based on the discretization of the vehicle evolution space into simple maneuvers, represented each by a simple dynamic model such as constant velocity or constant turning etc. This allows the method to be optimized for highly dynamic vehicles. Unfortunately classical IMM shows some drawbacks concerning some real time multi sensors applications. In this work, we focus on outdoor vehicle localization with asynchronous sensors in order to report these drawbacks and then propose a new solution. Many tests carried out with simulated and real data confirm the interest for using such a solution in our applications.