Autonomous navigation in complex environments with a micro-aerial vehicle

In this talk, I will discuss approaches that enable a quadrotor to autonomously navigate and explore complex indoor and outdoor environments. Micro-aerial vehicles (MAVs), and in our case quadrotors, offer exceptional 3D mobility over ground platforms, making them particularly suitable for search-and-rescue missions in which the vehicle must be able to navigate through complex 3D environments. In such missions, especially in response to emergency or disaster situations, it may not be safe for a human to enter the environment and therefore the MAV must be able to operate fully autonomously without requiring any human operator commands or external infrastructure. This talk summarizes a sequence of projects that move towards the goal of fully autonomous MAVs and will consist of three parts: (1) algorithms and systems design that enable autonomous exploration of complex indoor 3D environments with a quadrotor equipped with a laser scanner, an IMU, and limited computation; (2) a state estimation approach that permits autonomous navigation in mixed indoor and outdoor environments using laser, GPS and IMU; and (3) a vision-based state estimator that greatly expand the capable operational environments of our quadrotor platform. Extensive experimental evaluations are presented in each part of the talk.

Speaker: Shaojie Shen, Univ. of Penn

Room 531