A new hybrid stereovision-based distance-estimation approach for mobile robot platforms

In this paper, a new hybrid stereovision-based distance-estimation approach is proposed for indoor mobile robot platforms. To calculate the distances between the robot and object, triangulation method is used first. However, this technique is insufficient for objects located on the left or right side of the robot or the case of the stereo camera pair fixed to the robot with an angle down to the floor. Therefore, a new approach is proposed by adding a look-up table and curve-fitting methods to the triangulation technique. In this approach, the stereo camera pair is located on the robot with an angle down to the floor. In the experimental studies, an average accuracy rate of 97.69% is obtained. The Manhattan and Euclidean distances of objects that are not in the same line as the robot are also calculated. Average accuracy rates of 98.24% for Manhattan and 98.03% for Euclidean distances are achieved.