An Autonomous Flying Robot Maneuvers Easily Around Obstacles
An autonomous flying robot designed by Cornell researchers for search-and-rescue operations features ability to guide itself through forests, tunnels or damaged buildings. The base for the robot is a quadrotor, a commercially-available flying machine with four helicopter rotors. The quadrotors had been programmed to navigate hallways and stairwells using 3D cameras, but these cameras turned out inexact at large distances to plan a route around obstacles. At present assistant professor of computer science Ashutosh Saxena is working to turn a flat video camera image into a 3D model of the environment. The robot is being trained with 3D pictures of different obstacles like tree branches, poles, fences and buildings, learning the common characteristics of all the images, such as color, shape, texture and context. As a result of this, a certain set of rules for identifying obstacles is burned into a chip before the robot flies. During the flight the robot breaks the current 3D image of its environment into small chunks, determines the obstacles and calculates a path through them as close as possible to the set route, making necessary adjustments as the view changes.