DATA FUSION APPLIED ON AUTONOMOUS GROUND VEHICLES

Abstract Accurate environment sensing has become crucial in 2. Existent Work and Motivation various domains ranging from remote control, remote collaboration, navigation in autonomous vehicles and disaster prevention systems. The recent success of Stanford's Stanley's SUV in the The common characteristics for these applications are: diverse sensors operating at different rates resulting in the heterogeneity of DARPA [2] Grand Challenge Off-Road Robot Race proves data sources, constant processing time constraints, and real-time that a good software design on commercially available decision making processes based on the collected information. A vehicles far outstrips custom made solutions with reliable data fusion model must address each of these requirements inter-operability issues between the software layer and and must allow the user to constantly monitor the flow of data in the vehicle's hardware. framework. This paper proposes a hierarchical model for data fusion Autonomous driving is essentially a continuous environment using real time objects and an application program interface (API). sensing application producing continuous sensor data that needs to be integrated by the navigation module. Thi