Inter-vehicle sensor fusion for accurate vehicle localization supported by V2V and V2I communications

Cooperative driving system techniques aim to minimize accidents, traffic congestion and consequently the environmental costs of road traffic. An accurate vehicle's pose is of extreme importance for the inner working of the traffic management systems. An agent based traffic simulator was developed integrating typical automotive sensors, vehicle-to-vehicle (V2V) and vehicle-to-infrastructure (V2I) communications, vehicle's agent and infrastructure agent. In this paper, we propose to further enhance a multi-sensor localization algorithm in environments where V2V and V2I communication is possible. This paper describes an inter-vehicle sensor fusion for an accurate vehicle positioning, where each vehicle is modeled by an agent, and each agent provides information depending on its vehicle sensors. In the first fusion stage, data from four wheel encoders and one steering encoder are fused by means of an EKF, providing robust odometric information, namely in face of undesirable effects of wheels slippage. Next, a second fusion stage is processed for integrating odometric and inter-vehicle absolute positioning data.