Tech
Autonomous Flying Robots Can Avoid Obstacles
Brooke Miller
First Posted: Oct 31, 2012 04:57 AM EDT
An autonomous flying robot has been devised by the researchers at Cornell University. What makes this flying robot unique is that is can function as smart as a bird when it comes to maneuvering but there are still issues to be resolved.
The flying robot is designed with outstanding features such the ability to navigate through thick forests, tunnels or damaged buildings. It can be used in the search and rescue operation. Small flying machines are already common, and GPS technology provides guidance.
Assistant professor of computer science Ashutosh Saxena and his team are now dealing with crucial issue of how to keep the vehicle from slamming into walls and tree branches. It is difficult for the human controllers to react swiftly at all times, and it is impossible to receive continuous radio signals everywhere the robot flies.
The test vehicle is a quadrotor, a commercially available flying machine about the size of a card table with four helicopter rotors. The quadrotors have been programmed by the researchers to navigate hallways and stairwells with the help of 3D cameras.
The accuracy of these cameras diminishes in the wild and hence they are not apt enough at long distances to plan a route around obstacles. Due to this reason 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 unique flying robot was trained with 3-D pictures of various obstacles such as tree branches, poles, fences and buildings. This was initiated by graduate students Ian Lenz and Mevlana Gemici. 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. During the flight the 3-D images of the environment are broken down by the robot into small chunks based on obvious boundaries, decides which ones are obstacles and works out a path through them as close as possible to the route it has been told to follow, constantly making adjustments as the view changes.
This new design was tested in 53 autonomous flights in obstacle-rich environments including Cornell's Arts Quad and it succeeded in 51 cases, failing in two because of winds.
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.
The results were presented at the International Conference on Intelligent Robots and Systems in Portugal Oct. 7-12.
See Now:
NASA's Juno Spacecraft's Rendezvous With Jupiter's Mammoth Cyclone
©2024 ScienceWorldReport.com All rights reserved. Do not reproduce without permission. The window to the world of science news.
More on SCIENCEwr
First Posted: Oct 31, 2012 04:57 AM EDT
An autonomous flying robot has been devised by the researchers at Cornell University. What makes this flying robot unique is that is can function as smart as a bird when it comes to maneuvering but there are still issues to be resolved.
The flying robot is designed with outstanding features such the ability to navigate through thick forests, tunnels or damaged buildings. It can be used in the search and rescue operation. Small flying machines are already common, and GPS technology provides guidance.
Assistant professor of computer science Ashutosh Saxena and his team are now dealing with crucial issue of how to keep the vehicle from slamming into walls and tree branches. It is difficult for the human controllers to react swiftly at all times, and it is impossible to receive continuous radio signals everywhere the robot flies.
The test vehicle is a quadrotor, a commercially available flying machine about the size of a card table with four helicopter rotors. The quadrotors have been programmed by the researchers to navigate hallways and stairwells with the help of 3D cameras.
The accuracy of these cameras diminishes in the wild and hence they are not apt enough at long distances to plan a route around obstacles. Due to this reason 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 unique flying robot was trained with 3-D pictures of various obstacles such as tree branches, poles, fences and buildings. This was initiated by graduate students Ian Lenz and Mevlana Gemici. 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. During the flight the 3-D images of the environment are broken down by the robot into small chunks based on obvious boundaries, decides which ones are obstacles and works out a path through them as close as possible to the route it has been told to follow, constantly making adjustments as the view changes.
This new design was tested in 53 autonomous flights in obstacle-rich environments including Cornell's Arts Quad and it succeeded in 51 cases, failing in two because of winds.
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.
The results were presented at the International Conference on Intelligent Robots and Systems in Portugal Oct. 7-12.
See Now: NASA's Juno Spacecraft's Rendezvous With Jupiter's Mammoth Cyclone