AI image sensor slashes air conditioning energy

August 18, 2021 // By Nick Flaherty
AI image sensor slashes air conditioning energy
Nomad Go is using an AI-enabled image sensor from Sony to monitor rooms and cut air conditioning energy use by 30 percent

Nomad Go in the US has used the AI-enabled intelligent vision sensor IMX500 from Sony Semiconductor for a dynamic HVAC system that cuts energy use by 30 percent.

The system was developed with the support of the newly established Co-Innovation Lab operated by Microsoft and Sony and aims to help building managers reduce CO2 emissions and energy costs by over 30 percent in buildings.

Nomad Go’s AI-Power HVAC uses edge computer vision to control any HVAC system on-demand based on live occupancy and provides a customer-facing dashboard to provide real-time controls and insights about usage, energy, and CO2 savings. Sony’s IMX500 provides AI processing on the image sensor. This results in lower bandwidth use and addresses end-user privacy by enabling metadata output instead of sending the image data itself to the cloud.

Microsoft Azure Cognitive Services provides a scalable and adaptable computer vision service to enable large-scale, multi-building deployments of the Nomad Go solution.

Related articles

Commercial buildings account for 39 percent of global CO2 emissions, so reducing the energy usage is a major focus. One key reason for this is that HVAC systems in buildings are controlled by set schedules rather than by actual room occupancy. This results in long periods of heating, cooling, and ventilating empty spaces and leads to unnecessary energy expenses, decreased net asset value of buildings, and wasted CO2 emissions.

“Nomad Go’s edge AI solution combined with Sony’s intelligent vision sensor IMX500 are expected to be an excellent match for the real estate industry’s efforts to reduce greenhouse gas emissions while lowering energy costs,” said Yuki Tsuji, Deputy General Manager, System Solution Business Development at Sony. “Our IMX500 contributes to reducing bandwidth and latency, enabling more responsive and cost-effective HVAC operations. We look forward to ushering in more meaningful business opportunities like this for our customers and development partners.”

While traditional video analytics systems require expensive video hardware and high bandwidth costs, this solution reduces expenses across all areas.

“The Co-Innovation Lab experience is a highly effective way to ideate, design and implement powerful edge AI solutions through an expedited development process,” said David Greschler, CEO and co-founder of Nomad Go. “Our latest deployment for commercial and multi-family buildings quantifies how our AI edge computer vision platform contributed to a reduction of carbon emissions and energy costs by over 30 percent. What’s more, with direct access to resources and guidance from Sony and Microsoft through the lab program, we saw more than a 50 percent reduction in up-front costs compared to traditional AI edge solutions available in the market.”

“With Microsoft Azure AI capabilities embedded on Sony’s intelligent vision sensor, Nomad Go will be able to offer its customers a solution that will significantly cut down on CO2 emissions from commercial buildings,” said Rashmi Misra, head of AI, mixed reality and silicon business development, Microsoft. “By leveraging these cutting-edge imaging and sensing technologies for high-speed edge AI processing, Nomad Go’s customers will be able to manage their HVAC systems much more efficiently and sustainably.”

The Co-innovation lab provides independent software vendors (ISVs), developers and enterprise customers with the tools, knowledge, and global resources they need to create new smart camera solutions. Through its work in the lab, Nomad Go benefitted from complimentary technical and engineering support to more quickly ideate, develop, prototype, and test the solution in a short period of time.

Other articles on eeNews Europe


Vous êtes certain ?

Si vous désactivez les cookies, vous ne pouvez plus naviguer sur le site.

Vous allez être rediriger vers Google.