Localization, Planning, and Control of a UAV for Rapid Complete Coverage Bridge Inspection in Large-Scale Intermittent GPS Environments

In this article, we tackle the problem of automated visual inspection using a quadrotor unmanned aerial vehicle (UAV) in large-scale intermittent Global Positioning System (GPS) environments. There are mainly three main challenges in this task including drift-free localization in intermittent GPS environments, coverage planning for maximizing inspection efficiency, and safe flight in disturbance environments. An aerial system, comprising localization, planning, and control components, is presented to address the aforementioned challenges. In this system, a localization method called intermittent GPS-aided visual inertial odometry (IGVIO) is devised to achieve precision absolute localization under intermittent GPS by fusion of visual inertial odometry (VIO). A hierarchical time-optimal trajectory planning method called HTOP is also proposed to maximize the efficiency of complete coverage inspection, and finally, a nonlinear model predictive control (NMPC) scheme is leveraged to achieve robust trajectory tracking despite potential external disturbances. Compared with the state-of-the-art time-optimal planners, the proposed planner has key improvements in stability, coverage, efficiency, and optimality, which implies our proposal is more suitable for rapid complete coverage bridge inspection. The proposal is tested in real-world experiments with a commercial quadrotor, which demonstrates that our aerial system can meet the practicability demands of automated visual inspection.