The Evaluation of Navigation Capabilities Using Optic Flow Sensors on a Miniature Rotary-Wing Platform

This paper presents an evaluation of optic flow sensors for micro air vehicle (MAV) navigation and obstacle avoidance. Specifically, the use of optic flow information to estimate the distance from objects in the observed environment is investigated. The sensors under consideration output a single value estimate of optic flow representing the average optic flow in the entire field of view of the sensor. Two optic flow sensors were subjected to rigidly defined motion. The differences between the sensors, as well as their sensitivity to textures of varying contrasts and complexity, were studied. An equation is defined for calculating distance using a known calibration condition. This equation is then used to estimate distance to an observed scene with objects located at an unknown distance. Preliminary results indicate that the distance between a sensor and an object can be estimated with sufficient accuracy for MAV navigation and obstacle avoidance.