Lumentum Holdings Inc. and Ambarella, Inc. have developed a joint reference design that combines edge artificial intelligence with privacy for occupancy sensing in smart-building applications. The Vision-D reference design leverages Lumentum’s flood illuminator module for high-accuracy time-of-flight (ToF) 3D sensing and Ambarella’s CV22 edge AI SoC, which is said to deliver industry-leading AI performance per watt.
The combination of the Lumentum and Ambarella technologies allows small artificial-intelligence-of-things (AIoT) sensors with local processing to be deployed in buildings for applications such as occupancy monitoring, intelligent space management, and smart retail.
The CV22 SoC supports multiple image sensor inputs and offers a set of peripheral interfaces for a variety of sensors — including humidity, temperature, and audio — to enhance environmental perception. The Lumentum 10-W flood illuminator module uses a high-performance, three-junction vertical-cavity surface-emitting laser array for ToF 3D sensing.
The ToF 3D sensing maintains privacy by generating a point cloud with granular 3D information but without any personally identifiable information, which is suited for applications such as health care and eldercare, said the companies. Examples include remote monitoring of aging parents without invading their privacy, whereby AI algorithms running ON this reference design can monitor occupants for sudden falls or medical events and trigger a request for emergency services.
Vision-D reference design. Click for a larger image. (Source: Ambarella)
This is not the first time the two companies worked together on a reference design. Last year, Ambarella and Lumentum worked with ON Semiconductor (now onsemi) to develop two joint AI-based 3D sensing reference designs that are designed to accelerate AIoT device deployment. These reference designs — Saturn for electronic locks and Vision+ for biometric access control readers — build on the companies’ previous joint solution for contactless access systems.
The reference design is unique in that it combines sophisticated building automation and occupant monitoring with privacy, according to the partnership. “There are currently two choices for occupancy sensing in building automation systems: basic motion-detection devices that preserve privacy or intelligent camera-based systems that compromise privacy,” said Jerome Gigot, senior director of marketing at Ambarella, in a statement.
The joint reference design is expected to open up opportunities for adding intelligent sensing with AI processing at the edge, while offering privacy and eliminating the cost of cloud computing.
Integrating intelligence onto these sensing nodes allows functionality and scalability, said the companies, citing examples such as the creation of digital twins for automated real-time space utilization or hot-desking in shared workspaces. Other applications include traffic-based hospitality staffing, cleaning and maintenance alerts, and energy-efficient HVAC and lighting systems such as adjusting temperature based on real-time occupancy.
The Vision-D joint reference design is available now. Watch the video for a demonstration of anonymous occupancy sensing. Contact Ambarella for more information.