Unmanned Systems Technology 025 | iXblue DriX I Maintenance I UGVs I IDEX 2019 I Planck Aero Shearwater I Sky Power hybrid system I Delph Dynamics RH4 I GCSs I StreetDrone Twizy I Oceanology Americas 2019

35 around analysing performance data across UAS subsystems can increasingly detect signs of a potential failure. Lithium batteries are a notable example here, partly because they have to undergo repeated charge-discharge cycles to ensure they meet standards of quality. Their state of health and general performance during flights are also logged for analytical purposes. For example, battery analytics will note if significantly more current is used than is typical, or if a performance curve is tailing off. Neither the battery health system nor the technician would necessarily be able to pinpoint the root cause at that stage, but it still enables a sub-optimal battery to be pulled from service before a failure can occur. As analytics improve in their ability to connect performance variables with indicators of problems, scheduled preventive maintenance could be scaled back in favour of maintenance on an ad- hoc, predictive basis. That would mean lower UAV operating costs in terms of maintenance services and hours spent on the ground. It would also improve system lifespans by reducing the chances of human error, such as manually damaging wiring harnesses or screwing in fasteners too tightly, as it remains one of the biggest causes of accidents requiring repairs. However, the movement from preventive to predictive maintenance will depend on further advances in algorithms to automate the process of monitoring subsystems and notify users about issues that could render a UAV unusable. Somewhat paradoxically, however, human maintenance will still be required to identify human-caused issues as predictive maintenance advances. Performance monitoring sensors cannot spot a loose wire or improperly tightened fastener, unless some form of internal camera architecture is used. Service stations One thing that manned and unmanned aviation maintenance have in common is that the checks and troubleshooting of individual components tend simply to follow the guidelines or manuals provided by each component’s manufacturer. Among systems for unmanned vehicles, however, the absence of significant regulation means there is less commonality in these guidelines and practices. There are also almost never any manufacturer specifications on tolerances, mean time between failures or other information about how often a vehicle or part should be sent for servicing. As a result, it tends to fall to the in- house practices of service stations to develop their own MRO schemes and recommendations for the frequency and degree of maintenance. Times between preventive maintenance servicing can be determined as a function of several factors. Most important of these are the frequency with which UAVs fly, and the type of environments they fly in. Most organisations using commercial UAVs (to perform photogrammetry or surveys of civil infrastructure, for example) do not fly at rates requiring more than one visit to a servicing centre per year. If however such companies are flying in dusty environments, such as farms or construction yards, they might have their UAV(s) serviced every six months. There is a growing minority of commercial UAV operators though who currently (or plan to) fly multiple aircraft daily, regardless of precipitation, grit or temperature extremes. These can include delivery services, aerial mapping providers and power companies conducting daily surveillance and inspections of assets such as pipelines or power lines. For these companies, each UAV in their fleet should be serviced every two or three months to avoid a systems failure that could lead to the loss of an aircraft, as well as severe damage to an expensive payload or asset. It follows then that the bulk of servicing for commercial UAVs revolves around ensuring that all moving parts are clean and free from corrosion and its sources. For example, the rpm of brushless DC (BLDC) motors is such that even the smallest grain of sand can cause damage, so a key part of their servicing typically involves dismantling and cleaning them with pressurised air, an electrostatic brush or similar. Maintenance | Focus Unmanned Systems Technology | April/May 2019 Visual and manual checks of onboard systems make up the typical initial steps in maintaining unmanned systems (Courtesy of Fortress UAV)