A Localizability Constraint-Based Path Planning Method for Unmanned Aerial Vehicle

摘要

As unmanned aerial vehicles (UAVs) are used in challenging environments to carry out various complex tasks, a satisfactory level of localization performance is required to ensure safe and reliable operations. 3D laser range finder (LRF)-based localization is a suitable approach in areas where GPS signal is not accesible or unreliable. During navigation, environmental information and map noises at different locations may contribute differently to a UAV's localization process, causing it to have dissimilar ability to localize itself using LRF readings, which is referred to as localizability in this paper. We propose a localizability constraint (LC) based path planning method for UAV, which plans the navigation path according to LRF sensor model to achieve higher localization performance throughout the path. Paths planned with and without LC are compared and discussed through simulations in outdoor urban and wilderness environments. We show that the proposed method effectively reduces the localization error along the planned paths.