DARPA Quadrotor Autonomously Avoids Obstacles During Flight

Thursday, November 1, 2012

Flying robot drone avoiding obstacles
Quadrotor Robots
Cornell University researcher, Ashutosh Saxena and his team have created the algorithms to allow flying quadrotor robots to autonomously avoid obstacles.  The work is leading to flying robots that will act as birds do.
Researchers at Cornell University have developed an autonomous flying robot drone that maneuvering around obstacles as a bird would.  The project was funded by DARPA.

Guiding itself through forests, tunnels or damaged buildings, the quadrotor robot could have tremendous value in search-and-rescue operations.

Ashutosh Saxena, assistant professor of computer science, and his team are tackling the hard part: how to keep the vehicle from slamming into walls, tree branches and other obstacles. Remote human operators can't always react swiftly enough, and radio signals may not reach everywhere the robot goes.

Saxena and his team have 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.

robot autonomous flight path
View of the quadrotor's flight path calculated based on 3D model of the environment.
Image Source: Saxena Lab

Saxena is building on methods he previously developed to turn a flat video camera image into a 3-D 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 Ian Lenz and Mevlana Gemici trained the robot with 3D scans 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 -- a branch, for example, is attached to a tree. 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 robot was tested in 53 autonomous flights in obstacle-rich environments -- including Cornell's Arts Quad -- succeeding in 51 cases, failing twice because of winds. The results were presented at the International Conference on Intelligent Robots and Systems (IROS 2012) in Portugal Oct. 7-12.

Next, Saxena 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, he suggests having people throw tennis balls at the flying vehicle.

SOURCE  Cornell University

By 33rd SquareSubscribe to 33rd Square