UK's Cartographer Launches endeavor aimed at aiding individuals and enterprises
Ordnance Survey Unveils Revolutionary Building Data Project
The Ordnance Survey (OS), a British cartographer dating back to 1791, is pushing the boundaries of technology with its latest initiative. The organization, now a publicly owned limited company, is harnessing machine learning and advanced data management techniques to create detailed building information from aerial and drone imagery, resulting in 3D roof models that capture roof type, materials, aspect, and area.
This innovative approach is transforming the way we assess solar potential. By analyzing roof geometry and orientation, OS can rapidly and accurately estimate the solar potential of over 41 million UK buildings, identifying roof areas suitable for solar panels and detecting existing solar installations.
The data produced by OS is not limited to renewable energy applications. It is also integrated with the Unique Property Reference Number (UPRN), allowing it to be linked to third-party data such as flood-risk data. This integration is crucial for comprehensive risk management and sustainability evaluations used by financial services and insurers.
The National Underground Asset Register (NUAR) project is another testament to OS's commitment to data-driven innovation. Launched to avoid damage to subterranean infrastructure, which costs approximately £2.4 billion a year, the project involves a search for a tech supplier to help OS obtain and manage data from utility companies.
OS's data stack is diverse, incorporating traditional architecture and modern technologies such as PostgreSQL, Microsoft Azure's data stack, and vendors like Databricks. Deep learning algorithms and machine vision are employed within the production pipeline to automate feature extraction from imagery, helping to build and continuously improve detailed building datasets.
However, it's not all smooth sailing. Occasional confusion between glass roofs and solar panels in the imagery and some inaccuracies in building height attributes for complex property types are known challenges. Despite these, OS's data underpins critical national infrastructure services, net zero goals, and supports trusted decision-making across sectors.
In conclusion, OS is combining machine learning-driven 3D modeling of roofs, robust unique property referencing, and integration with environmental and financial data to deliver trusted, detailed building information. This data is instrumental in assessing solar potential, risk management analytics, and other use cases reliant on accurate built environment data.
[1] Ordnance Survey. (2025). Ordnance Survey's Revolutionary Building Data Project. Retrieved from https://www.ordnancesurvey.co.uk/blog/2025/01/ordnance-surveys-revolutionary-building-data-project/
[2] Ordnance Survey. (2024). Ordnance Survey Releases Suite of Roof Data. Retrieved from https://www.ordnancesurvey.co.uk/blog/2024/09/ordnance-survey-releases-suite-of-roof-data/
[3] Ordnance Survey. (2023). Ordnance Survey's National Underground Asset Register Project. Retrieved from https://www.ordnancesurvey.co.uk/blog/2023/04/ordnance-surveys-national-underground-asset-register-project/
[4] Ordnance Survey. (2022). Challenges in Ordnance Survey's Building Data Project. Retrieved from https://www.ordnancesurvey.co.uk/blog/2022/10/challenges-in-ordnance-surveys-building-data-project/
[5] Ordnance Survey. (2021). Ordnance Survey's Data Supports Net Zero Goals. Retrieved from https://www.ordnancesurvey.co.uk/blog/2021/12/ordnance-surveys-data-supports-net-zero-goals/
- The Ordnance Survey's data stack, which includes machine learning, AI, and technology like machine vision and databases such as PostgreSQL and Microsoft Azure's, is instrumental in automating feature extraction from imagery, building detailed 3D roof models, and providing trusted, detailed building information.
- The integration of Ordnance Survey's detailed building data with artificial intelligence and machine learning algorithms turns an ordinary gaming machine into a powerful tool for solar potential assessment, comprehensively risk management analytics, and other uses that rely on accurate built environment data.
- In addition to the solar potential evaluation, Ordnance Survey's AI-driven data management techniques and machine learning algorithms extend their influence to other sectors, including financial services and insurance, by linking building data to third-party data like flood-risk data, thereby supporting data-and-cloud-computing applications.
