Development of Real-Time Lane Detection Software for Automated Steering System

In recent years, there has been a growing interest in the development of Intelligent Transportation System (ITS), providing safety and comfort to passengers, and thereby decreasing the number of accidents. In addition, the efficiency of highway utilization can be increased tremendously by replacing some of the basis tasks of driving such as accelerating to a pre-specified velocity, braking within a safe distance, and automatically controlling the steering movement of the car within the left and right lanes with the help of intelligent vehicle control systems. One of the intelligent vehicle control system components is the automated steering system. Generally, the automated steering system can be further categorized into two major subsystems, the route detection and the control system. The development of these subsystems involves multi-disciplined areas such as instrumentation, signal and image processing, mathematical modeling, and control engineering. There are two approaches in developing the route detection system, the look-down and the look-ahead. This research focuses on the development of the route detection system based on the look-ahead approach where a machine vision system is used as the detector for the automated steering system to extract lane information.