Integrated localization system for autonomous unmanned aerial vehicle formation flight

This paper presents a formation control framework based on an integrated position sensor system, including two cameras and several inertial sensors, for Unmanned Aerial Vehicles (UAVs) to fly outdoors. A sliding mode controller provides the desired orientation for the UAV to track a desired trajectory. The flight control system takes into account the estimated attitude and position to guide the follower vehicle along trajectories that maintain adequate formation stability. The cameras are employed to detect the relative positions by extracting feature points from specific markers on the UAV. The features are processed to estimate the attitude and position of the follower vehicle. A real-time guidance algorithm is also proposed to merge image measurements with Inertial Measurement Unit (IMU) data for state estimation. The system has been successfully tested in flight experiments, which verify the effectiveness and applicability of the proposed control framework and visual measurement algorithms.