Toward autonomous navigation using an RGB-D camera for flight in unknown indoor environments

Because of their potential important role in tasks of 3D space which ground vehicles cannot perform, autonomous flight and exploration of micro aerial vehicle(MAV) has attracted increasing attention. Many researches focused on flight with RGB cameras or laser range finders, whereas MAV with RGB-D cameras has been rarely mentioned. In this paper we present an autonomous navigation system for indoor flight with only onboard sensors - an RGB-D camera and a low-cost MEMS inertial measurement unit(IMU). In order to get the best performance of navigation, many visual odometry techniques have been evaluated through experiments. Then, hover and trajectory flight experiments prove the design is valid and efficient for autonomous flight in complex indoor environments.