Rule-based tracking of multiple lanes using particle filters

Tracking of lanes is essential for intelligent vehicles in order to drive autonomously. A system is presented which allows tracking of multiple lanes. The system is based on a clear modelling of a lane and the parameter set of each lane is estimated using a particle filter which fuses different cues. A finite-state machine then decides whether or not a lane is really tracked. For each lane, a separate tracker is used and a set of rules controls the life-cycle of all trackers and keeps track of all the estimated lanes