Steps to build a DIY low-cost fixed-wing drone for biodiversity conservation

drone developed in this project as low-cost since the summed amount of the components was $995, substantially less than the average value of commercial drones, even in relation to other DIY drones used in biodiversity studies [See S1 Table]. The reduction of drone acquisition costs in conservation projects produces a significant saving in the total budget, but there are additional factors to consider when evaluating the DIY option against commercial products. While the material cost of the drone developed here is $14802 less than commercial drones’ average, it took us five months to develop it and an additional month to perform the tests pre- sented in this study. Among the few studies that describe the development of a low-cost fixed- wing drone in the last decade [9, 18, 19, 27, 28] there is no information on the average develop- ment time, which makes it impossible to compare our results with other studies. This fact makes the development time informed here a useful parameter for the development of future projects. Therefore, when choosing between developing your own drone system and purchas- ing a commercial RTF system, the time required for development must be considered in addi- tion to the final cost. As in other studies that use low-cost fixed-wing drones to carry out conservation works [4, 15, 16, 19], we also chose for this project the development of the drone system of the glider- type platform due to its better aerodynamic efficiency, portability and competitive cost.

Despite the choice of this platform, the development process is not limited to the specific plat- form model presented here. Once the main components and the way to make their connec- tions and configurations are defined, these can be easily mounted into other fixed-wing platforms without many adjustments or even translated to rotary-wing platforms, although with some more modifications. While the development of the first DIY model requires a sub- stantial initial time investment, thereafter only occasional updates are necessary for building new DIY platforms, even for different goals. Through simple transect and lawnmower flights tests, we verified that our model served for monitoring large areas within a 20 km radius, covering more than 1 km2 in a single flight at high spatial resolution, which is sufficient to perform standard vegetation analysis [29, 30], fauna identification [10, 31] and deforestation monitoring [2]. Particularly for deforestation monitoring, the ability of this model to flight long distances, enabling large coverage with high spatial resolution, makes it a low-cost technology tool with a great potential for combating ille- gal activities in protected areas, especially in the Cerrado biome, one of the world’s biodiversity hotspots [32] which is suffering a drastic loss of native vegetation during the last years [33]. The maximum flight time and coverage capacity of this model was similar to that of com- mercial RTF systems such as the standard version of eBee model (SenseFly), one of the most

well-known fixed-wing systems in the drone market, and which presents an average acquisi- tion cost 15 times higher than the model developed in this project. Beyond the economic fac- tor, other important factors, still little considered in the comparisons between the use of drones and manned airplanes, are the environmental impacts and the social costs concerning greenhouse gas emissions. One of the first attempts to analyse these costs showed that moni- toring areas of up to 30 km2 using photometry with a resolution of 5 cm / px from fixed-wing drones is more economically and socially advantageous than the use of manned aircrafts [34]. Considering that the comparisons in that study were made between the costs of manned air- craft and the eBee model, representing the fixed-wing drones, the economic advantage of DIY drone models as the one developed in this project is emphasised. Although nowadays there are satellite systems such as DETER [35] used in Amazon and Cerrado biomes monitoring with a greater potential to identify changes in forest vegetation cover in areas measuring between 25 and 100 ha, with a spatial resolution of between 56 and 64 meters, the system developed here can identify changes in vegetation cover at a scale of meters and with spatial resolution at a scale of centimetres. By conducting only two PLOS ONE Building a DIY low-cost fixed-wing drone for biodiversity conservation PLOS ONE | https://doi.org/10.1371/journal.pone.0255559 August 13, 2021 13 / 17