Visual Looming as a Range Sensor for Mobile Robots
MetadataShow full item record
This paper describes and evaluates visual looming as a method for monocular range estimation. The looming algorithm is based on the relationship between displacements of the observer relative to an object, and the resulting change in the size of the objeet's image on the focal plane of the camem. Though the looming algorithm has been described in detail in prior reports, its usefulness for inexpensive, robust ranging has not been realized widely. In this paper we pwpose that visual looming can be used to extract ranging information much as with sonar. The accuracy of the looming algorithm is found to be significantly more robust than sonar when the object whose distance is being measured is slanted relative to the robot's line of sight. On the other hand, sonar is better suited for objects that cannot be easily be visually segmented from their background, or objects that do not fit entirely within the focal plane. We demonstrate these results using a camera and frame grabber mounted on a Pioneer 1 mobile robot. Our results suggest that the looming algorithm is an excellent method to complement sonar for traditional ranging techniques.