Dynamical reconfiguration strategy of a multi sensor data fusion algorithm based on Information Theory

Autonomous vehicle board an increasing quantity of sensors such as GNSS receivers, road maps, LIDARs, to ensure a reliable localization function. This expanding number leads to abandon a classical centralised data fusion method and to adopt more complex architectures such as Distributed Data Fusion. In this paper, we propose an approach to detect faults of the different sensors in such configuration and to dynamically reconfigure the fusion method by using simple concepts of Information Theory. In order to detect faults, consistency is examined through a log likelihood ratio between the information innovation of each sensor. The current study illustrates the performance of the proposed fault detection algorithm and the pertinence of the dynamical reconfiguration of the multi-sensors data fusion. Experimental results, using data from antilock braking system sensors, a differential global positioning system receiver, and an accurate digital roadmap illustrate the performance of the proposed approach.