Improved sensing and planning for enhanced flight duration allow unpiloted small aircraft to fly autonomously in cluttered environments over long distances. Can we engineer an artificial “homing pigeon”—that is, create a small aircraft that can perform a task for us, pilot itself, and travel for a long time over great distances? Uninhabited aerial vehicles (UAVs), which have been developed mainly for military applications, are still remotely controlled, and some tasks, such as flying in crowded environments, are too difficult, dangerous, or expensive to be performed by even the smallest of these aircraft (1). These problems have spurred the development of smaller robotic air vehicles. Specifically, micro air vehicles (MAVs) are inexpensive, fly autonomously, and are small and lightweight (with wingspans of 15 cm or total mass of 0.5 kg) (2). We focus on the two technical challenges that must be met by MAVs if they are to fly as well as birds, namely perception—sensing and responding to their environment—and persistence—staying airborne.
[1]
Sebastian Thrun,et al.
Learning Metric-Topological Maps for Indoor Mobile Robot Navigation
,
1998,
Artif. Intell..
[2]
J. Ashby.
References and Notes
,
1999
.
[3]
G. Sachs.
Minimum shear wind strength required for dynamic soaring of albatrosses
,
2004
.
[4]
Guillermo Rein,et al.
44th AIAA Aerospace Sciences Meeting and Exhibit
,
2006
.
[5]
Arthur Richards,et al.
Engineless unmanned aerial vehicle propulsion by dynamic soaring
,
2009
.
[6]
Nicholas Roy,et al.
Stereo vision and laser odometry for autonomous helicopters in GPS-denied indoor environments
,
2009,
Defense + Commercial Sensing.
[7]
Ingemar J. Cox,et al.
Dynamic Map Building for an Autonomous Mobile Robot
,
1990,
EEE International Workshop on Intelligent Robots and Systems, Towards a New Frontier of Applications.