Cornell University researchers have created an autonomous flying robot that is as smart as a bird when it comes to maneuvering around obstacles. Able to guide itself through forests, tunnels or damaged buildings, the machine could have tremendous value in search-and-rescue operations. The team is studying how to keep the vehicle from slamming into walls and tree branches. Human controllers can't always react swiftly enough, and radio signals may not reach everywhere the robot goes.
The test vehicle is a quadrotor, a commercially available flying machine about the size of a card table with four helicopter rotors. The team has already programmed quadrotors to navigate hallways and stairwells. But in the wild, current methods aren't accurate enough at large distances to plan a route around obstacles. Researchers are building on techniques previously developed to turn a flat video camera image into a 3D model of the environment using such cues as converging straight lines, the apparent size of familiar objects and what objects are in front of or behind each other -- the same cues humans unconsciously use to supplement their stereoscopic vision.
Graduate students trained the robot with 3D pictures of such obstacles as tree branches, poles, fences and buildings; the robot's computer learns the characteristics all the images have in common, such as color, shape, texture and context. The resulting set of rules for deciding what is an obstacle is burned into a chip before the robot flies. In flight the robot breaks the current 3D image of its environment into small chunks based on obvious boundaries, decides which ones are obstacles and computes a path through them as close as possible to the route it has been told to follow, constantly making adjustments as the view changes.
The team plans to improve the robot's ability to respond to environment variations such as winds, and enable it to detect and avoid moving objects, like real birds; for testing purposes, people will throw tennis balls at the flying vehicle.