A thesis submitted in partial fulfillment of the requirements for the degree of Master of Engineering, School of Engineering and Technology

Thesis (M.Eng.) - Asian Institute of Technology, 2006

Transportation has undergone a great number of beneficial breakthroughs, especially in automatic applications. The intelligent vehicle which features an automatic steering system is one of the most important. The thesis tried to develop a vision-based automatic steering control system for an intelligent vehicle. In order for a car to steer itself automatically, it must know its position with respect to the road (road mark). A machine vision lane detection algorithm was developed by processing the gray image from a CCD camera to detect the road mark. A program was written in Microsoft visual c++ 6.0 to capture and process the image dynamically. This program was used to obtain the lateral position and offset angle of the car's center line relative to the road mark. A microcontroller received the output from a computer through serial communication and computed the steering angle to control a servomotor to drive the car along the desired road. A prototype was implemented to verify the algorithm