If you’re looking for startup ideas that can help slow climate change, you might become an expert in home energy assessments. At least, that was the case for the founders of Kelvin, a French startup that uses computer vision and machine learning to facilitate energy efficiency audits of homes.
Clémentine Lalande, Pierre Joly and Guillaume Sempé started to look at energy efficiency audits of homes, because renovation works have a huge impact on reducing energy consumption and CO2 emissions. But like other companies in the construction industry, most companies in this sector do not use technology to improve their processes.
“300 million homes are scheduled to be renovated in Europe over the next 30 years,” Kelvin CEO Lalande told TechCrunch, “and yet the construction industry is the second least digitally adapted after agriculture.”
In France, the National Agency for Housing (ANAH) has set itself the ambitious goal of renovating 200,000 homes in 2024 alone. But craftsmen are struggling to keep up, worsening the environment as a result. More generally, the regulatory environment in Europe is favorable for this kind of startup.
Founded in October 2023, Kelvin is a pure software company: The company isn't looking to build a marketplace for service providers, nor does it want to become a consumer-facing product, unlike Enter, another Germany-based home energy rating startup covered by TechCrunch.
Instead, the startup assembled a small team of engineers to use machine learning to create its own AI model focused on home energy assessments. The company uses open data, such as satellite imagery, as well as its own training dataset, which includes millions of photos and energy assessments.
“We calculate over a dozen proprietary, semi-public or open data sources that provide information about buildings and their thermal performance. So we use fairly standard segmentation techniques and analyze satellite imagery with machine learning models to detect certain features like the presence of neighboring buildings, solar panels, collective ventilation units, etc.,” Lalande said.
“We also do this with the data we collect ourselves. We've developed remote inspection tools with bots that tell people inside what photos and videos they should collect,” she added, “and we have models that count the number of radiators in a video, detect doors, detect ceiling heights, determine the type of boiler or ventilation system.”
Kelvin doesn't want to use 3D technology such as LIDAR because he wants to build a tool that can be used on a large scale: you don't need the latest smartphone with a LIDAR sensor to record the details of a room, as regular photos and videos can be used.
Potential customers for the startup could include construction companies, the real estate industry, and even financial institutions looking to fund home improvement projects, especially as lenders may want an accurate valuation before making a decision.
In the company's initial tests, the Home Energy Rating was accurate to within 5% of the older rating, and if the rating becomes the go-to tool for these audits, it will make it much easier to compare homes and renovations to each other.
The startup has raised 4.7 million euros ($5.1 million at today's exchange rates) to date, with Racine² leading the round and a non-dilutive investment from Bpifrance. Seedcamp, Raise Capital, Kima Ventures, Motier Ventures and several business angels also participated in the round.
Image credit: Kelvin
Source link