Icing and wind – implications and mitigations in high-intensity, safety-critical drone operations in Norway

The background of this project is a desire at the Oslo University Hospital (OUS) to explore the feasibility of using drones to transport blood samples between two major hospitals in Oslo. The motivation is to achieve improvements in patient care, and possibly consolidate laboratory services to a single location. We start by discussing the user requirements, based on a thorough study performed at OUS. We conclude that there is a need for frequent drone flights (possibly 15 minutes apart), carrying relatively light loads – under 3.5 kg. Safety is paramount. We continue by describing how current commercial or professional small drones are affected by icing and wind. Relevant weather statistics are summarized, and analyzed in the context of the use case. We discuss possible mitigations, and attempt to outline essential elements in the path towards making the vision of drone-based transport a reality. We especially underline the importance of developing new solutions for planning and control, leveraging a high level of automation. This is necessary e.g. to make good use of meteorological services in route planning and decision-making. We also emphasize strongly the requirement for ice protection systems (IPS) on board the drones. Furthermore, we identify the need to study the responses of drones to icing and strong wind in greater depth. A comprehensive and holistic approach should be taken. Solutions should be gradually matured based on what is learned in low-intensity, risk-controlled exploratory operations. We finally conclude that transporting blood samples with drones between Rikshospitalet University Hospital and Ullevål University Hospital is realistic within 5–10 years, provided emerging IPS and new planning and control technology are used. A residual risk of serious incidents and accidents will remain, however, even if the suggested approach is followed. This risk stems from rare weather events, as well as from other threats to safe operations such as collisions and technical malfunctions. Risk acceptance and mitigation should be studied further.