Innovation and talent at Libelium: the key to cleaner air
At Libelium, we believe that innovation is built on talent. Therefore, within the framework of the Programa Investigo 2022 and with the support of the Plan de Recuperación, Transformación y Resiliencia, Libelium Lab has developed an ambitious R&D&I project: the application of Artificial Intelligence models to improve the accuracy of air quality sensors. Pedro Antonio Sánchez is the leader of this project and his work is fundamental in the development of our environmental monitoring solutions
AI to improve accuracy in air quality monitoring
Pedro Antonio has joined our team to lead an essential project: the “Development of AI models to improve the accuracy of air quality sensors”. We understand the importance of having reliable data in environmental monitoring, especially for cities and companies committed to public health and sustainability. This is what Pedro likes most about this project: “What motivates the most is the feeling that what you do is going to have a real impact, is going to be used and is going to have a use for people. “
The challenge is clear: air quality sensors can be affected by factors such as temperature, humidity or variations over time. Let’s say they can be a bit… on their own! Pedro Antonio’s work focuses on optimizing the reliability of these measurements through the use of Artificial Intelligence (AI). His approach involves deep analysis of historical data to identify patterns and correct deviations. Through the implementation of deep neural networks and the use of technologies such as Python and TensorFlow, advanced algorithms are being developed that are capable of adjusting and calibrating sensor readings automatically. This process ensures that the information collected is always accurate and consistent.
The learning process has been continuous, as Pedro relates: “At the beginning we wanted to focus too much on solving the final section, and we lost sight of the process to be carried out. We have been rectifying, carefully observing the whole process, and we realized that there was room for improvement in certain processes even before applying AI”.
From research to application
Pedro Antonio’s project is not limited to the research phase. It seeks a real impact on environmental monitoring to generate practical and scalable solutions. To this end, we are developing a service accessible through APIs, which facilitates the integration and exchange of data with other company systems. The goal is to make our devices and solutions interoperable and easy to implement in different environments.
“For me it has been an opportunity for growth, both personally and professionally. An opportunity to consolidate and expand my knowledge, to participate in exciting and demanding projects and to work with exceptional colleagues,” says Pedro Antonio.
Validation of these models is performed through rigorous testing with equipment identical to that which will be deployed in the field. These evaluations ensure that the systems perform optimally in real conditions, and that the accuracy of the data generated is compared and validated with reference stations.
The result is tangible: a significant improvement in the accuracy of our sensor measurements. This translates into more reliable, real-time data for our users, enabling more informed and effective decision making in air quality management.
At Libelium, the incorporation of talents such as Pedro Antonio, supported by initiatives such as the Programa Investigo, reaffirms our commitment to responsible innovation and the creation of solutions that contribute to healthier and more sustainable communities.
