Analysis and experimentation of an optical-flow-based navigation algorithm for a lander

This paper deals with the development and the testing of an algorithm for the landing phase of a space vehicle, based on the optical-flow technique to measure the trajectory’s velocity. The Optical-flow sensor is obtained through the processing of 2-D images captured by a camera by knowing the acquisition time for each frame. Image processing techniques have been used for the implementation of an algorithm capable of detecting complex structures. The algorithm can detect craters and the shadows created by the craters themselves. It is possible to obtain the direction of the light by simply connecting the center of the crater with the center of the shadowed zone. The information of the light direction is used to obtain a solar compass. Finally, an extended Kalman filter has been used to estimate the state. The navigation filter is tested on the Lunar Simulator Facility at the Automation, Robotics, and Control for Aerospace Laboratory (ARCAlab) of the Department of Astronautical, Electrical and Energy Engineering of Sapienza University of Rome.