Drone4u

The objective of "Drone4u" is to bring new capabilities to drones (a.k.a. UAVs), in particular to low-cost drones such as the ARDrone from Parrot. A few capabilities we are targetting are described below.

The autonomous navigations of drones inside buildings is currently our main target. Drones usually navigate by going from one GPS position to another one. That kind of navigation is obviously restricted to areas where the GPS signal is strong, and where the needed position precision is around a few meters. Therefore, GPS-based navigation cannot be used inside buildings. To navigate inside buildings, we have proposed vision-based algorithms in order to automatically understand the environment, so that drones can take navigation decisions on their own (= autonomous navigation).

Many new applications could take advantage of drone autonomous navigation in our society: disaster assistance (exploring buildings on fire, finding an available route for safety teams after earthquake/flooding), daily-assistance (guiding persons in large building complex, assisting disabled people), etc.

The followings video illustrates the autonomous navigation of a drone:

• Following a red line located on the floor
• Recognizing signs/landmarks: the drone can analyze signs so as to take decisions at important points, e.g., turning left/right at a crossing.
• Interpreting the environment in 3D so as to avoid obstacles, and also to minimize landmarks (e.g., avoid putting a red line along the drone paths).

Another important issue we intend to deal with is people / group tracking, as illustrated by the following video.