Development of UAV Localization and Navigation Techniques for Warehouse Inventory Inspection

The popularity and availability of UAV technologies have made industrial automation more flexible and easy to adopt. In this paper we develop a UAV navigation system for warehouse inventory inspection application. Based on the SLAM framework, a fusion scheme by integrating UWB positioning is proposed to achieve accurate and robust localization. In addition to the global planning, an obstacle avoidance algorithm based on fuzzy control and VFH is developed to generate locally smooth trajectories. The proposed techniques are verified with simulation and experiments in real-world environments. Using a commercially available drone modified with depth sensor and single-board computer, the results have demonstrated the feasibility and effectiveness of our method for collision-free path planning and obstacle avoidance in various application scenarios.