Inventory management applications using unmanned aerial vehicles (UAVs) appear to have the highest potential for use in warehouse operations. They remove the need for manual inspection, which is in most cases performed in heights using ladder, pen and paper. Moreover, it provides a competitive edge in the field of logistics while decreasing costs and improving stock management efficiency. Deployment of autonomous UAVs in an indoor environment puts higher requirements for the level of autonomy due to the more complex nature of the environment (narrow crossings, absence of Global Navigation Satellite System (GNSS), smaller manoeuvring space, dynamic environment, etc. ...). In this paper, we would like to introduce a system for automation of planning, managing, and executing a warehouse inspection. The goal is to provide an easy-to-use framework of the inspection task that any warehouse operator is capable of using it. The system consists of two main parts - user interface and path planner. Using a created human-machine interface (HMI), an operator is able to define the workspace and the inspection task. The path planning module takes as an input the definition of the inspection task defined by the operator and plans an optimised collision-less path. The operator can visualise the generated mission, launch it, observe the mission execution progress in real time and receive the mission results. The functionality of the system is verified by a simulation as well as by test flights made in an environment simulating a warehouse by a real UAV called AV Discovery built by Airvolute s.r.o.
Science Journal 