A System of Robot Detecting and Tracking Moving Objects Based on Modified Optical Flow

Robot’s adaptability to surrounding as an essential subject of robot research is based on detecting environment precisely and acting timely. Moving object detection and tracking is an important approach of robot detecting external environment. Moving object detection and tracking based on computer vision means building a system with vision part and motor drive of the robot. In this paper, the robot estimates the motion of object and tracks it by driving motors according to the changes of optical flow fields got from modified optical flow calculation of collected images. In the experiment, this system responded quickly, stably and accurately. And it could make the robot move exactly and timely to respond to changes of surrounding.