Lane tracking for on-road targets

Considerable research has been done on tracking ground targets, including on-road targets. Lane tracking of an on-road target is a new problem in the ground target tracking area. Knowledge of the lane that a target is located in is of particular interest in on-road surveillance and target tracking systems. This paper proposes a method to track the lane of an on-road target based on a Hidden Markov Model (HMM). It is assumed that an image sensor provides raw observation data. The formulation of the problem and our approach are given. Simulation results show that the implemented HMM-based lane estimators can achieve good performance.