Semi-autonomous teleoperation of UAVs in search and rescue scenarios

Unmanned Aerial Vehicles (UAVs) exhibit great agility and usually require a trained pilot to operate them, while being restricted to line-of-sight range. This prevents their wide introduction in applications such as inspection for disaster scenarios. Our goal is to enable rescue teams untrained in piloting to teleoperate a UAV as an aid to their mission, while being able to focus on the task at hand. The assisted teleoperation of a small UAV is presented in this work, within a framework to perform localization and state estimation of the UAV without any external positioning system (e.g. GPS) and perform autonomous local obstacle avoidance. The computational efficiency of the whole processing pipeline allows for real-time operation. Field experiments have been conducted in a realistic mock-up using an indoor testbed with a very accurate tracking system in order to validate the algorithms.