Distance estimation using stereo vision for indoor mobile robot applications

Real time and accurate estimation of the distance between the robot and the object or the distance between two different objects is critical for robot applications. Various sensors as laser, sonar or camera have been used for this process. Two sensors as a camera and laser are employed together in some studies in literature. Camera based studies propose different methods using single camera or two cameras. In this study, the distance between the cameras and the colored object (red colored object is accepted in this study) is calculated in pixels and cm using two cameras situated in the same axis. In this proposed study, the images taken from the left and right cameras are preprocessed first. Then, the objects in the environment are determined. The last process is to estimate the required distance. 150 measurements, including distances between 60 cm and 140 cm, are performed in the experimental studies and an accuracy success over %90 is achieved.