Automated Vehicle System Using Both A Computer Vision And Magnetic Field Sensors

Abstract

For automated vehicle highway systems, a reliable sensor which measures vehicle position and road curvature is necessary. This paper describes an integrated sensing system which consists of a computer vision and magnetic field sensors with inductive cable. Especially, a new image processing technique based on a template matching algorithm is developed. Not only vehicle position measured by magnetic field sensors but also knowledge of road structures is taken into account to detect lane boundaries. Road curvature can be estimated accurately from the detected lane boundaries and vehicle yaw angle. Experiments were carried out and provided encouraging resulu.