Vision based Collaborative Localization for Swarms of Aerial Vehicles

We present a framework for localizing a swarm of multirotor micro aerial vehicles (MAV) through collaboration using vision based sensing. For MAVs equipped with monocular cameras, this technique, built upon a relative pose estimation strategy between two or more cameras, enables the MAVs to share information of a common map and thus estimate accurate metric poses between each other even through fast motion and changing environments. Synchronized feature detection, matching and robust tracking enable the use of multiple view geometry concepts for performing the estimation. Furthermore, we present the implementation details of this technique followed by a set of results which involves evaluation of the accuracy of the pose estimates through test cases in both simulated and real experiments. Our test cases involve a group of quadrotors in simulation, as well as real world flight tests with two MAVs.