The Autonomous Intelligent Systems (AIS) Group at the University of Freiburg in Germany has designed an indoor robotic helicopter (quadrotor) with a high degree of autonomous behavior. Unlike mobile robotics platforms with wheels, a system framework for flying platforms equipped with sensor systems was not available, until now. The AIS group has made their quadrotor system design and software framework available as open source under the Open Quadrotor Project. Xsens has supported the research of the AIS group as part of the EU-project MuFLy.
The AIS group has achieved a remarkable degree of autonomous behavior. The small helicopter can simultaneously create a map of the environment and localize itself (SLAM) using a scanning laser ranger stabilized by an MTi-G IMU/GPS as well as track and map altitude. Some very nice videos of the quadrotor in action can be found here, including examples of dynamic obstacle avoidance.
These kind of flying platforms can be used for reconnaissance in areas that human rescue workers cannot access, or are too dangerous to enter, such as unstable buildings and structures after an earthquake.
The paper related to this research, published at ICRA’09, won the Best Conference Paper Award. [Slawomir Grzonka, Giorgio Grisetti, Wolfram Burgard, Towards a Navigation System for Autonomous Indoor Flying, IEEE International Conference on Robotics and Automation (ICRA), 2009]
