Autonomous Exploration in Unknown Urban Environments for Unmanned Aerial Vehicles

In this paper, we present an autonomous exploration method for unmanned aerial vehicles in unknown urban environment. We address two major aspects of explorationmap building and obstacle avoidanceby combining model predictive control (MPC) with a local obstacle map builder. An onboard laser scanner is used to build the online map of obstacles around the vehicle during the flight. A real-time MPC algorithm with a cost function that penalizes the distance to the nearest obstacle replans the path. The adjusted trajectory is sent to the position tracking layer in the Berkeley UAV avionics. The proposed approach is implemented on Berkeley rotorcraft UAVs and successfully tested in urban flight experiment setup.