Abstract
Nowadays, an increasing number of assistances is available to help the driver and to improve the vehicle handling. However, most of these driving assistances only have few interaction with the driver. Moreover, each assistance acts separately, and only aims to reduce the risk of the situation while keeping the vehicle on the lane. With increasing sensing capacities, it now becomes possible to have a local view of the vehicle surrounding. Hence, the next step is to define a new layer that is able to interact with the driver, accordingly to the knowledge of the environment. This layer then will be able to determine the maneuver that must be applied, the haptic feedback for the driver and supervises actions of driving assistance. In this article, we propose a decision method that aims to evaluate the risk related with the traffic and determine the best set of maneuvers.
Keywords: risk, collision avoidance, maneuver planning