During our latest mission, we went to Ethiopia to map malaria hotspots. The idea is to look for small water bodies which in the dry season are a source for mosquitos and their larvae. From these hotspots, the mosquitos spread in the rainy season all over the region, and if infected by the malaria parasite can do a lot of damage to the local community. I have done several demonstration flights and trained local students on how to operate the UAV.  During one mission we can map an area of 2km2 on a resolution of 7cm. The plane is an USENSE-X8 running APM2.5 with 10000Mah 3S batteries on a 900kv motor and a 13*8 prop. The camera is a canon s100, running CDHK which we trigger via USB. The imagery were processed using a Pix4D discovery license, which is free, and which create wonderful fly-throughs. I am especially happy with the way Pix4D is processing the 3D structure of vegetation. Much better than other SfM software I have seen thus far.

Now the students are processing all the imagery and looking into performing these flights on their own. This project is part of a series of development projects funded by the Flemish government (VLIR-UOS) where universities of Belgium collaborate with institutes in the South to exchange knowledge with each other. 

In the framework of the VlIR-UOS-North-South-South initiative aimed at capacity building for developing countries, Latin-American and African Universities participate in a 7 day workshop on UAV-technology, currently being held in Lima, Peru. In this workshop organized by University Nacional la Molina, researchers from Peru, Ecuador, Colombia, Cuba and Ethiopia receive training on UAV mapping, making use of low-cost and open-source technologies. Beside the theoretical training and hands-on with simulators and training planes, also a 3-day demonstration is organized in the Andes region. With the UAV imagery, the researchers hope to study and improve productivity of small scale agro-ecosystems. In the next month’s 4 other demonstrations will be organized at the participating countries, ranging from sugar cane monitoring in Cuba to fighting malaria spread in Ethiopia.

The training is given by my company, Usense, which is using the APM 2.5 as autopilot for mapping purposes. I am happy to say that the researchers were very enthusiastic to work with this open-source technology and found it very user-friendly. You'll probably see some of them around here in at Diydrones in the coming months. Currently I am giving them a demonstration with the USENSE-X8 UAV on high altitude. Yesterday we flew at 4000m, with some stunning views from the Andes! 

Prof. Eddie Schrevens (KULeuven), who is leads the project: “We have been investing quite some money on satellite imagery to evaluate the productivity of agro-ecosystems in many parts of Peru, without success due to high cloud coverage in these regions. Since one year we are doing tests with imagery taken from a UAV resulting in high-quality orthomosaics and digital surface models. These UAVs fly, and fly well, but now we need to investigate in smart algorithms to derive the right information from the imagery.

Some more pictures from the workshop

University grounds: Above

Below you can find some results of a 4 week mapping campaign I did in Peru along with people from KU Leuven, Belgium and University La Molina, Peru. Images were processed with PIX4D and dronemapper. Special thanks to JP from Dronemapper for the good support and tips&tricks. I hope you like the videos!

Archeological site (Caral):

Amazone area:

During a 4 weeks mission to Peru we have done several UAV flights for mapping purposes. These included flights at high altitude over the Andes. The highest one was over the scenic “Bosque de Piedras”, at 4500m altitude, where we have mapped a 1km2 area in a matrix flight pattern.

 - Images processed with PIX4D - 

These flights fit within a Belgian - Peruvian cooperation (KU Leuven – UNALM) to study and improve agricultural production in coastal, Andes and Amazon areas of the country. Special thanks to all the developers here at diydrones, dedicated to make Arduplane rock-solid and performing well under difficult conditions. During the 30+ autopilot missions, Arduplane never failed  and we managed to finish the project with one X8 frame, despite the harsh conditions (rainy season, high altitude, challenging landing spots..). Below a road-movie from the Andes part of the trip: