The global edge AI market size was valued at USD 14.7B million in 2022 and is expected to expand at a compound annual growth rate (CAGR) of 21.0% from 2023 to 2030 according to Grand view research. The edge AI market is based on the on-edge computing as a main pillar. Embedded edge computing involves processing data closer to the source of generation, reducing the need to send all data to centralized cloud servers. Running machine learning models directly on endpoints like cameras, sensors, and other IoT devices, instead of relying solely on cloud-based servers has several challenges to overcome such as power and cost. AI acceleration in endpoints can provide low power and flexible solution to support newly developed AI models.
Power efficiency and optimal image processing for Vision AI can be reached with Dynamically Reconfigurable Processor for AI inference (DRP-AI I). DRP-AI’s good power efficiency can eliminate the need for dealing with heat dissipation solutions such as heat sinks or cooling fans. Another advantage is that DRP-AI supports both real-time AI inference and image processing functions as well as camera support such as color correction and noise reduction instead of external image signal processor (ISP).
Renesas RZ/V2L SoC is equipped with a Cortex®-A55 (1.2GHz) CPU. The integrated DRP-AI includes DRP and AI-MAC in addition to 16-bit DDR3L/DDR4 interface and a built-in 3D graphics engine with Arm Mali-G31 and video codec (H.264). Renesas RZ/V2L DRP-AI servs as cost efficient AI accelerator with high-speed AI inference processing that provides low power and flexibility based on the reconfigurable processor ideal for AI Inference.
SolidRun partnered with Renesas to offer a RZ/V2L SoM, a System on a Module (SoM), also known as a Computer on a Module (CoM), is a compact, integrated computing solution that encapsulates key components such as processor, memory, storage, and various other components on a modular board. SoMs offer faster development cycles as designers can focus on the application-specific features and functionality without the need to design the entire system from scratch.
The RZ/V2L System On Module features:
- Dual Core Arm Cortex-A55, 1 x Cortex-M33, 1.2GHz.
- Onboard Certified Dual Band WiFi 802.11 a/b/g/n/ac with Bluetooth 5.0
- Built-in DRP-AI accelerator, its Pin to Pin compatible with RZ/G2L application processor SoM.
One of its SW key advantages is the CIP (Civil Infrastructure Platform) Linux that ensures robust, sustainable infrastructural system.
The SoM Form Factor, Mezzanine, size 47 x 30 mm, is offered in Commercial (0°C to 70°C) and Industrial (-40°C to 85°C) grade. Power Consumption is smaller than 5W and its supported OS is based on Embedded Linux (Yocto/Debian Distribution). In addition to evaluation Boards, SolidRun offers industrial grade ready to deploy SBCs with customization options.