Artificial Intelligence, Internet of Things and Digital Transformation. Never has been as much talk about technological innovations as nowadays. The importance of adapting to and following changes that the world demands is increasing every day, mainly in the business world.
Among the thousands of possibilities which Artificial Intelligence (AI) provides for the world there is Machine Learning, a term which refers to a technology in which computers are capable of learning and responding based on an analysis of different types of data.
Artificial Intelligence allows us to find information within those analysed data. As a matter of fact, if they are not properly employed, those data alone have no usefulness whatsoever. Therefore, Machine Learning is present in a number of forms in our day-to-day life, such as, for instance, in the use of some applications and at the countless sites for online shopping.
An analysis made by McKinsey in the year 2018 with more than 400 use cases showed that Artificial Intelligence is superior to traditional analytics techniques in 69% of potential use cases. Furthermore, the same analysis proved that applications of AI can be seen in all sectors of economy and various business functions, from marketing to supply chain operations.
If those disruptive technologies absolutely reach all markets, with the space and agriculture sectors it could not be different.
In an interview for Forbes Insider, Cleber Oliveira Soares, Director of Innovation at the Ministry of Agriculture, Livestock and Purveyance (MAPA, in Portuguese), spoke about digitalization of agrobusiness. For him, the trend is that, in a post-pandemic world, the agrobusiness sector undergo deep transformations, accelerating digitalization of the whole chain. “This digital transformation involves acceleration of adoption of a series of state-of-the-art technologies. Efficient data management, through Big Data, for instance, is capable of improving information on inputs, climate conditions and machinery to the extent of lowering costs and increasing productivity”, declared Cleber.
Technology applied to solution of global problems
For 2100, the Earth is predicted to have a total population of 10.8 billion people. Such a growth is inversely proportional to the quantity of resources we shall have in order to sustain that situation.
Hypercubes — one of the startups we invest in — was born inside Singularity University, in the Silicon Valley, and develops a technology which allows nanosatellites to photograph the surface of the Earth in order to analyse the soundness of the soil, with the objective of ceasing scarcity of resources for the next generations.
According to data from the Ministry of Agriculture, Livestock and Purveyance, in the last decades the productive chain of the agriculture sector has produced goods and services which amount to R$ 1 trillion per year, including inputs, agricultural production, agroindustry, and services related to the activity. Undoubtedly, technology can still further increase and potentiate investments in the sector.
Specifically focused on agrobusiness, Hypercubes places the producer in the heart of that science. However, from the very beginning of its course, the founders of the startup perceived a huge challenge to be faced: Technology. Artificial intelligence and the hardware where it trafficked along became democratic, and technologies came to be tested rather easily. The great difficulty was in testing those innovations in space.
The enterprise was created in the beginning of 2015, but it was not until the year 2017 that the first version of the technology they needed to make data processing inside the very satellite, almost in real time, came to be commercialized. “With democratization of the technology, we found the other half necessary to make possible remote processing. Without that, we would take weeks to download data from each satellite; we calculated that each satellite generates 100 terabytes of data in every 90 minutes”, told us Fábio Teixeira, founder of Hypercubes.
Machine Learning in guaranteeing resources to the next generations
One of the main competitive differentials of Hypercubes is concentrated in Machine Learning, since the enterprise produces customized models for identification of parameters which detect problems such as, for example, diseases, anomalies and soil stresses of farming.
The algorithms of Machine Learning affect, rather much, the images from satellites. As a matter of fact, they allow an active monitoring which builds intelligent information, impressed from correct data, delivered in the right time, for the right people, and for an adequate price.
As its very name spells, the process employed by Hypercubes works fully based on the learning by machines. Within its properties, agriculturists can teach computers about all anomalies that might arrive at the soil of their farming. Following that, the learned information returns to space, where satellites start to photograph the surface of the Earth in search of what has been taught them by the producer.
“One thing is taking a picture to observe a problem, but when you process those data in the satellite itself, you can program the constellation in order that every time it flies over [a given soil], it searches for that already known problem. We leave [aside] a passive monitoring and go for an active monitoring”, explains Fábio Teixeira.
This whole process of loaded hardware, software and technology provides that problems are detected in antecedence, at their initial stage. Thus, plantations come to offer better resources, in a more sustainable manner and in larger quantity, bringing Hypercubes closer still to its primary objective: Putting an end to hunger in the world.
“When we place in orbit a constellation of satellites capable of daily taking pictures and processing data, we have everything that machine learning needs — it’s a treat for artificial intelligence”, concludes the founder of the startup.