MIT develops autonomous plane
August 13, 2012, 01:57 pm
Research teams recently competed in a series of autonomous-helicopter challenges from the Association for Unmanned Vehicle Systems International, and progress has been so rapid that teams are beginning to step it up. The last two challenges were done indoors, without the use of GPS.
Massachusetts Institute of Technology's Robust Robotics Group, which won the competition last year, is ramping up competition and developing autonomous-control algorithms for the indoor flight of airplanes.
It has been a goal of the team to carry the autonomous technology over to airplanes. At the 2011 International Conference on Robotics and Automation, MIT researchers explained an algorithm for the autonomous plane's trajectory. In 2012, during the same conference, the team presented the algorithm for the plane's states, location, physical orientation, velocity and acceleration. The team has now successfully completed various flight tests, in which the autonomous plane and its state-estimation algorithm was successful navigating its way around pillars and other obstacles in a parking garage at MIT's campus.
“The reason that we switched from the helicopter to the fixed-wing vehicle is that the fixed-wing vehicle is a more complicated and interesting problem, but also that it has a much longer flight time,” said Nick Roy, an associate professor of aeronautics and astronautics and head of the Robust Robotics Group. “The helicopter is working very hard just to keep itself in the air, and we wanted to be able to fly longer distances for longer periods of time.”
Roy added that the plane is far more complicated because it goes fast, but can't do arbitrary movements, go sideways and hover.
The plane is provided a digital map of its environment, unlike the helicopter which have to draw maps as they go. Nonetheless, the autonomous plane has to determine its location on the map in real time, using data it receives from a rangefinder and inertial sensors, accelerometers and gyroscopes. In order to determine its location, it has to calculate 15 different values, including how much it's tilted in any direction, its velocity and its acceleration.
“Navigation of lightweight, dynamic vehicles against rough prior 3-D structural maps is hard, important, timely and, I believe, will find exploitation in many, many fields,” said Paul Newman, a professor of information engineering at the University of Oxford and leader of Oxford’s Mobile Robotics Group.
With the amount of progress the autonomous plane has made in the last year, the team's next goal is to develop algorithms that can build a map while the plane is in the air.
Pilots should consider life insurance, even if the day comes where they won't have to touch the controls.
Are you covered? Are you overpaying? Find out! Get a Quote Now!