Autonomous Flight for Detection, Localization, and Tracking of Moving Targets With a Small Quadrotor
2017
with Justin Thomas, Giuseppe Loianno, Kostas Daniilidis, and Vijay Kumar
In this work, we enable a small quadrotor to autonomously track a moving target and plan trajectories that allow the robot to converge on the target. Our approach respects the dynamic, sensor, and actuator constraints of the vehicle, in particular ensuring that the target does not leave the field of view of the camera, which is rigidly attached to the robot and therefore affected by the coupled dynamics of the system. One of the major contributions of this work is that all sensing and computation is done onboard the 250 g vehicle, with no outside assistance from motion capture systems or external processors.