Expert Controller for Road Tracking Mobile Robot

the road boundaries provide useful information for performing safe vehicle moving paths in intelligent or expert vehicles. Many previous researches have treated road boundary detection, using different types of sensors such as vision, ultrasonic, RADAR, LIDAR (Light Detection and Ranging), and LRF (Laser Range Finder). In spite of that these sensors show advantages for road boundary extraction including good resolution but IR sensor is the most famous device for measuring small distances since it has a high accuracy, fast response time, low cost and a wide field of view. However, none of the previous works examined the problem of detecting road boundaries when roads could be structured with narrow regions. In this paper, we developed road boundary detection and tracking algorithm using two IR sensing for structured roads with narrow regions distributed along the road. The algorithm extracts road dimensions and distances relative to Left and Right IR sensors keeping the robot vehicle in the center of the road. The measured distances are kept constant by controlling analog inputs of the sensors using an expert controller. ARM processor controller is used to control the navigation of the mobile robot in this environment. The speed and direction of the robot are controlled by Pulse Width Modulation (PWM) signals generated by the ARM controller. Emphasis has been attributed to the design of the motor driver switching module in order to achieve effective and energy saving driving using power transistor drivers. The system is tested using a road prototype made from wood prepared for that purpose. It proved a good working in environments with narrow regions yielding a safe movement for the robot.