Free Space in Front of an Autonomous Guided Vehicle in Inner-City Conditions

This paper deals with the driving environment perception of cybercars from a vision system. The planification of the trajectory for an Autonomous Guided Vehicle requires to robustly segment the free space around the vehicle to manage an obstacle avoidance task. Opposite to the highways conditions, the urban environment still remains as the most complex environment due to the static scene can not be easily modeled (from urban canyon to large multi-lanes boulevards) and the variety of dynamic elements which animate the scenes do not follow the same path and have their own ranges of speed. The feature-based approach we detail consists on the segmentation of the ground surface according to the reduced field of view due to the traffic and the spatiotemporal constraints imposed by the tracking of features in several views.