Intelligent Visual Tracking Robots Using Cascade PID Controllers

A tracking system is useful for a variety of robotic applications, including obstacle detection, autonomous path planning and surveillance. One of the most difficult issues in computer vision is visual object tracking, which involves estimating the position of a target in an image series. Utilizing color information for visual tracking poses a formidable challenge due to variations in illuminant, shadows, shading, specular highlights, cameras, and object geometry. Consequently, color measurements exhibit significant fluctuations throughout an image series. A camera is frequently used for object tracking to acquire information on construction robot motion, which may subsequently be used in coordination controls. The paper presents a mobile robot vision system that allows the robot to recognize and track objects based on their color. A new approach is proposed for stable tracking of mobile robot using combined algorithm based on computer vision techniques and a cascade proportional integral derivative controllers (PID). Within a system designed for controlling the speed of a DC motor, the goal of a PID controller is to maintain speed at a specified reference value while dynamically accepting new set point values. The combination of feedback encoder and PID controllers can be employed to control motor speeds in the presence of noise and load disturbances