Introduction

The UNICEF Venture Fund is proud to see portfolio member, Rentadrone, graduate. They’ve come a long way – from numerous product iterations to deep diving into understanding their ecosystem better, strengthening their business model, and gearing up to take their solution to market. They’re now ready to collaborate at a larger scale – as they find new pathways to work with partners, investors, and the open source community.

Rentadrone uses machine learning, thermal, and multispectral imagery to detect crop diseases and monitor solar power plants Improving food security and energy efficiency of sustainable energy sources. During the course of the investment period, Rentadrone developed machine learning algorithms to detect and classify failures in solar plants using aerial thermal imagery  (single images and thermal orthomosaics). Using drones to capture these images, we have mapped 230,000 solar panels, detecting faults in over 7000 panels with the key objective  to increase energy efficiency and improve performance of solar farms. 

Along with this use case, we also deployed drones to take multispectral images of crops, using these images to develop algorithms that are able to determine possible health problems which after we will have to reconfirm by taking some samples and sending it to the lab for deeper investigation, analysis and diagnosis. Based on the historical calculus and analysis of vegetation indexes, we can establish, for example, if a crop has poor vigor, weak plants or not adequate relations among the quantities of leaf, flower and fruits. Currently, we have the goal to form partnerships with producers to collect much more multispectral drone data of their crops to test our algorithms. This way, we can have the feedback from the field to measure the accuracy of our models. With this additional use case, our team aims to contribute to precision agriculture.

Prototyping

We have two examples of the way we used to present our reports to clients in 2018, and how we present our results now using our own online platform. These two examples were made in the same photovoltaic farm, property of Engie, Los Loros. 

Imagine that you’re looking for problems in the grass of an enormous field, the size of one hundred football fields. If you do it by yourself, by walking these fields one-by-one, you might pass some problems in the grass, and it will take a lot of time and effort. The solar farms are pretty similar, they are giant solar farms, and to address the failures manually is too expensive and slow. This is where drones take charge. Drones are made to see things in a perspective that humans can’t, and this perspective is capable of covering more terrain, in a fraction of the time. 

2018 Example Photovoltaics (OLD REPORTS)

This was the way we presented the reports in the past, before participating in the UNICEF Innovation project, we used to present our reports in a very non-intuitive way, very linear and not too friendly.  Shown Below:

User/Field testing

During the investment period our team conducted thorough user testing; the most critical and extensive testing was done in  Pampa Camarones and El Aguila. These photovoltaic farms are in the middle of the Atacama Desert in Región de Arica y Parinacota at the north of our base and home country Chile. These two solar farms have friendly testing facilities and are commonly visited by junior engineering students and postgraduate students who want to explore different parts of the renewable energy sector. 

El Aguila was used by Engie as a technology testing lab, so there you can find different kinds of photovoltaic generation. This was the first solar farm to have photovoltaic generation in Chile, and they also have a photovoltaic concentrated solar power plant. Through this phase our team came to appreciate the need for continuous technical improvement and testing. Both solar farms, because of their size, helped our team to test in a controlled environment. 

Engie is a global energy company that has a presence in 70 countries worldwide, and they have contracted us to inspect their solar assets in Chile. 

Open Source

Open-source has facilitated and given scope to our product development. First,  it has provided the option to receive feedback faster and very specialized (if you have the right contributors), and that feedback often had a direct impact on the behaviour and features our product has now, and how our product and service should look in the years to come.

Finally, developing in the open has allowed us to share our knowledge about drone data  with communities and entrepreneurs all over the world who can implement  both algorithms — one for photovoltaic solar panel detection, and the other (OpenAgro) to detect health status among different kinds of crops. 

Business Models

At the onset of the engagement with the UNICEF Innovation Fund, we had a different approach about pricing. Now, with the help of all the consultants UNICEF has provided us, we have a broader approach about how to sell our product as a Software as a Service (SaaS), bringing on more volume-focused sales within our portfolio and future clients from different regions ranging from Asia, Europe and Latin America. 

Challenges

Our biggest challenge - and our biggest goal - is internationalization. We can’t keep that idea out of our heads, because we know we have an amazing product, and it’s our duty to take care of that and take our technology to international markets and compete with the best companies in our industry.  We know that taking the company and our partnerships to the next level takes sacrifice, but we trust in our capabilities and the potential of our business to grow faster; however, this requires having the right conditions, like access to the capital we need and the right management. 

Future collaborators

We’re looking to collaborate with computer vision and machine learning companies that could add more value to our current product. Having alliances with companies with high technical knowledge in data science for image processing would help us validate our procedures, allowing us to develop new quality processes making more precise predictions. We are also keen to collaborate with companies in the renewable energy sector so as to align needs to what challenges are being faced in that market and address them with our expertise. 

Next set of goals

We have inspected 2 million solar panels so far. We are looking forward to operating at least in four different countries this 2021 and also adding another 4 million solar panels to our count!

Working with the UNICEF Venture Fund

UNICEF has been vital in many ways. Aside from the funding, the assessment we’ve received from Justin Flory about open-source development, and Bernino Lind in market development, has been really useful for us as a company to set clear goals of where we see ourselves in the next 5 to 10 years. 

Feeling endorsed and supported by good professionals is really valuable and important when you’re trying to make things happen, and most importantly when you’re trying to innovate.

Another interesting development for us during the investment period is the way we’re developing business relationships with other companies within the cohort. We’re working closely with qAIRA and Dronfies Lab in trying to close new deals in their home countries and getting them involved in the operational part of the work that has to be done. It is always good to have friends and colleagues in different countries — these connections will be really useful for all of us. For example, if some of the companies decide to move on with a go-to-market in one of the base-countries from a known company, we could share advice and tips with each other making that landing softer