IA4Birds: A convergence of technologies and methodologies
The IA4Birds project brings together different technologies and methodologies to address the pivotal challenge of sustainable natural resource and energy management. With a focus on research into information fusion techniques to achieve the mass storage of bird count and transect data, as well as their activities and flight characteristics, direction, and height. The project comprises a research and development process that encompasses a number of key stages, each of which contributes significantly to the construction of an efficient and effective platform.
In this regard, the use of NoSQL technology; a standard API and a knowledge base are noteworthy.
Selecting the appropriate NoSQL technology: One of the key pillars of the IA4Birds project is the choice of the most appropriate NoSQL technology (databases specifically designed for specific data models that have flexible schemas for creating modern applications) for the integration of the data repository. Given that data from birds and external sources, such as protected areas and wind resources, can vary in structure and storage requirements, a thorough evaluation of different NoSQL database systems has been carried out, including options such as Cassandra, MongoDB, among others. Each of these technologies has been analysed in terms of its ability to handle heterogeneous data and scalability, ensuring that the final choice fits the functional requirements of the platform. Finally, the choice of MongoDB was based on our previous experience and its ability to meet the functional requirements of the platform.
Development of a common or coordination API: To ensure interoperability and efficiency in inserting and querying data in the repository, a common API (Application Programming Interfaces) has been designed and developed. This API allows CRUD (Create, Read, Update and Delete) operations to be carried out seamlessly and also supports bulk data query operations. Such abstraction of data access significantly simplifies the process of integrating various data sources and makes it easier to perform advanced analyses on the collected information. In addition, in our approach we added two other APIs managed by this common API: an ingest API, for acquiring data from different sources, including information captured through camera and microphone monitoring via the IoT equipment, and a third AI API, responsible for running the models needed for the application. If real-time data processing is required, we evaluate the feasibility of using a topics system such as Kafka or Socket, depending on the scale of the project. This choice will be made with the objective of guaranteeing an efficient and agile management of the information in real time.
Knowledge base deployment: The knowledge base resulting from the IA4Birds project has been effectively implemented for the storage of bird data and external sources. This includes the extraction of data from prohibited areas and protected areas of the Junta de Castilla y León, the acquisition of bird locations through the E-bird platform, as well as the collection of wind resource data from the Iberian Wind Map and the location of wind turbines currently installed in Castilla y León. This strategic deployment of the knowledge base is essential to optimise the assessment of sites for the implementation of wind farms and to ensure the adequate conservation of bird populations and surrounding ecosystems.
The IA4Birds project, coordinated by the AIR Institute, is funded by the Biodiversity Foundation of the Ministry for Ecological Transition and Demographic Challenge (MITECO) in the framework of the Recovery, Transformation and Resilience Plan (PRTR), funded by the European Union - NextGenerationEU. Its objective is to use artificial intelligence in combination with audiovisual devices to monitor bird populations and thus gain a better understanding of them, helping to prevent possible threats and to decide whether a site is suitable or unsuitable for the construction of a wind farm.