The Positive Impact of AI in Earth Observation
ESA’s Rochelle Schneider: AI’s Role in Earth Observation and Society
In the rapidly evolving field of Artificial Intelligence (AI), the European Space Agency (ESA) is leading the way with their upcoming Φsat-2 mission. The integration of AI in Earth observation has raised both hopes and fears, but Rochelle Schneider, the AI Applications Lead at ESA’s Φ-lab, sets the record straight on the positive impact of this transformational technology.
Schneider defines the role of AI in society as using computers and machines to automate highly repetitive tasks, allowing humans to focus on activities that require emotional intelligence, human relationships, and intuition. While there are concerns about job loss, Schneider believes that AI will create new opportunities and should be seen as a tool to imropve human abilities rather than replace them. ESA is actively teaching young scientists and engineers to embrace AI technologies in fields like Earth observation, integrating machine learning into their understanding of data and data processing.
One of the significant contributions of AI in Earth observation is in computer vision. Traditional algorithms or models used to identify specific objects in satellite images have limitations, as they cannot adapt and learn from new data. However, with AI, the model continuously evolves and learns as new data arrive, revolutionizing automatic detection. With the enormous volume of Earth observation data produced daily, automation driven by AI becomes an essential tool for monitoring the planet.
Schneider highlights two significant projects where AI has played an important role. In the Child Connectivity project with Giga, AI algorithms are used to recognize school buildings from satellite data to assess young people’s access to the internet. This initiative identified 65,000 incorrectly located schools in Brazil alone, using Copernicus Sentinel-2 data. In collaboration with UNICEF, AI-driven climate-data models were able to forecast dengue fever outbreaks one month in advance.
The rise of AI has also impacted the commercial space sector. The ESA InCubed program has witnessed exponential growth in commercial Earth observation companies in Europe. These enterprises utilize AI-powered tailoring of satellite data to provide real-time information to government departments, disaster response agencies, and insurers. The upcoming Φsat-2 mission is equipped with onboard AI processing that will eliminate clouds from images, create street maps, and detect shipping. Furthermore, commercial companies will have the opportunity to upload and run their own applications on the satellite, opening up new business models and commercial opportunities.
Looking to the future, Schneider envisions AI playing an important role in Earth observation by integrating with other emerging technologies. The Destination Earth project aims to create an AI-driven digital twin of the planet to monitor natural and human activity and anticipate extreme events. Schneider goes even further, envisioning a digital twin of Earth in the metaverse, providing immersive experiences through Extended Reality technology.
Overall, AI tools are set to become mainstream in ESA’s operations and Earth observation activities. They will play a significant role in building a green and sustainable society by providing insightful information for decision making and the stewardship of the planet. With ESA’s pioneering work in AI for Earth observation, the future looks promising as we embrace the power of this transformational technology to improve our understanding of the state and future of our planet.