Aerial Grasping by a Quadrotor UAV Based on a Soft Gripper

Despite the potential need for autonomous package delivery by drones, aerial grasping is a challenging problem due to various issues such as the ability to identify an object, navigate towards it, as well as to grasp various objects effectively. The research investigates autonomous grasping by a quadrotor UAV by utilizing vision based tracking on top of GPS-based navigation system and integrating a gripper made of soft material to handle various types of objects. Based on real-world experiments, the UAV’s flight stability, object detection ability and the performance of the soft gripper to grasp various objects are analyzed.