Agriculture is one of the oldest and most crucial fields of industry that is continuously developing. Year by year, global demographic growth means that demand for food is constantly increasing. Farmers need detailed information about soil parameters to cultivate their crops properly, so they require a system that will significantly speed up and facilitate this.
KP Labs and QZ Solutions have initiated a truly innovative project based on artificial intelligence to collect information on soil parameters using onboard machine learning techniques to analyse hyperspectral data to enhance agricultural processes.
This work is funded by the European Space Agency (the GENESIS project) and supported by the ESA Φ-lab. The GENESIS project developed by KP Labs and QZ Solutions aims to improve this process using the Intuition-1 satellite. The KP Labs team is striving to develop an Earth observation system based on using a satellite equipped with a hyperspectral sensor and onboard data processing capabilities using deep neural networks. Hyperspectral imaging collects and processes information from the electromagnetic spectrum. This means that the camera is able to scan the biochemical composition of soil and provide an overview of all ingredients present. The satellite will be launched by KP Labs by the end of 2022 and will collect data on soil parameters such as potassium, phosphate, magnesium, and pH levels. The traditional approach to monitoring soil parameters is completely human-dependent, as soil samples must be collected, mixed, and sent for analysis, which takes about 3 weeks and is very labour-intensive. By comparing the traditional method with the presented technology that will be on board Intuition-1, it will be possible to reduce the time to get results to 4 days, resulting in a significant improvement in field operations.
There are already many satellite remote sensing systems on the market whose photos can be used in agricultural analyses. Unfortunately, their size makes sending them back to Earth expensive and protracted. Thus, in the GENESIS project, the data will be processed onboard Intuition-1 with the help of the “Leopard” data processing unit, so the volume will be reduced, as only parameter maps will have to be transferred. This feature distinguishes this system from others, making the process much faster and decidedly less costly.
“Genesis will not only contribute to agriculture but also prove that machine learning can deliver key insights from raw hyperspectral data in specialized applications – making a great leap into the future of farming,” says Michal Zachara, COO of KP Labs.
If you are interested in the details of our solution or if you want to learn more about the mission we are planning, we will do our best to answer your questions. Contact us at BD@kplabs.space