Artificial intelligence (AI) vision silicon company Ambarella (Santa Clara, CA), optics and photonics products maker Lumentum (Milpitas, CA), and CMOS image sensor solutions provider ON Semiconductor (Phoenix, AZ) have developed two new joint reference designs that accelerate Artificial Intelligence of Things (AIoT) device deployment, building on the companies’ previous joint solution for contactless access systems. By combining the data from Lumentum’s vertical-cavity surface-emitting laser (VCSEL) array illuminators and an ON Semiconductor image sensor using Ambarella’s AI system on chip (SoC), higher levels of accuracy and more intelligent decision-making can be achieved in next-generation AIoT devices for biometric access control, 3D electronic locks, and other intelligent sensing applications.
Originally intended for biometric access control and electronic locks, these new reference designs can also address the needs of smart cities, smart buildings, smart homes, and intelligent healthcare. Additionally, the high level of integration provided by these joint solutions significantly lowers system power consumption and thermal design requirements while enabling much smaller product form factors.
The three companies’ new joint AIoT solutions consist of two reference designs and additional 3D sensing development kits, each configured to address application-specific demands for the combination of AI processing, 3D depth sensing, and vision sensing:
- Targeted at next-generation biometric access control readers, the Vision+ reference design is a 4K solution to perform single-camera, 940 nm structured-light sensing based on red-green-blue infrared (RGB-IR) technology. It leverages a single-chip solution for depth processing, AI processing, and video processing. Based on the Ambarella CV22 CVflow AI vision processor, it includes single-camera structured-light sensing powered by Lumentum’s VCSEL technology, featuring a range of 2 m, and a 4K (8 Mpixel) RGB-IR CMOS image sensor from ON Semiconductor.
- Targeted at next-generation smart electronic door locks, or eLocks, for both commercial and residential applications, the Saturn reference design integrates AI processing for both single-camera structured-light sensing and fast-boot video processing. It is based on the Ambarella CV25 CVflow AI vision processor and includes a structured-light camera powered by Lumentum’s VCSEL technology and the AR0237CS 2MP RGB-IR image sensor from ON Semiconductor.
- Targeted at intelligent sensing applications and with opportunities across verticals, Ambarella’s CV2 series CVflow AI vision processor-based development kits can now be outfitted with time of flight (ToF) sensor adapters that are powered by Lumentum’s VCSEL arrays. An optional 4K RGB-IR image sensor adapter is also available for these kits.
The open software development kit (SDK) for Ambarella’s CVflow AI vision processors allows integration of third-party applications while enabling OEMs to address differing regional requirements with a single platform (for example, using different AI algorithms in different regions). Additionally, this Linux-based SDK is said to expand the capabilities of edge sensing devices, simplify product development, and speed time-to-market through:
- Custom application development capabilities
- Support for different 3D modalities, including structured light and ToF. A single Ambarella SoC provides 3D sensing processing without the need for a dedicated depth processor or a separate host processor
- High AI inference capabilities on the camera device to enable multiple neural networks running simultaneously
- AI tools to support deep neural network development and migration
- Integrated ISP to handle challenging scenes, including high dynamic range and low-light scenarios
- Built-in Arm processors to run customer applications
- Integrated security hardware, including secure boot, OTP to store encryption keys, and Arm TrustZone technology for secure encryption key comparisons
- Rich peripheral interface support to simplify system designs
The joint reference designs and development kits are expected to be available in June 2021. For more information, contact Ambarella.