SolidSense AIoT - An IP64-rated AI vision development platform with Renesas RZ/V2M or Hailo-15 System-on-module

SolidRun SolidSense AIoT is an AI vision development platform designed based on the company’s Renesas RZ/V2N or Hailo-15 system-on-module with 15 TOPS and 20 TOPS of AI performance, respectively.

The solution comes with up to 8GB RAM, 256GB eMMC flash, WiFi 5, Bluetooth 5.0, and LTE Cat 1bis connectivity, an 8.4MP Sony IMX678 camera, four IR LEDs, and a few sensors all housed in an IP64-rated enclosure for outdoor operation. The SolidSense AIoT platform takes three 18650 rechargeable batteries with USB-C charging and an optional solar panel. There’s limited information about the Renesas model for now, so we’ll focus on the Hailo-15 variant in the rest of the article.

SolidSense AIoT Hailo-15H specifications:

System-on-Module – SolidRun Hailo-15 SoM
- SoC – Hailo-15
  - CPU – 4 x Cortex A53 @ 1.3GHz; 12 kDMIPS
  - Neural Network Core – 20 TOPS
  - VPU – 4Kp30 H.265/4 video encoder
  - ISP – 12MP
- System Memory – Up to 8GB LPDDR4 (4GB by default)
- Storage
  - 8 to 256 GB eMMC flash (32 GB by default)
  - Bootable 32 Mbit QSPI flash
- Wireless Connectivity – Dual-band 802.11a/b/g/n/ac WiFi 5 + Bluetooth 5.0
- Supply Voltage – 5V
- Dimensions – 47 x 30mm
Storage – MicroSD card
Camera – 8.4MP Sony IMX678 image sensor with 4x IR flash LEDs
Audio – Speaker and ultrasonic microphone (SPH 0641)
Wireless
- WiFi/Bluetooth on SoM
- LTE Cat 1bis connected over USB 3.1, Nano SIM card slot
Sensors
- Ambient Light Sensor
- Temperature and Humidity Sensor
- Optional PIR motion sensor
Power Supply
- 5V Via USB-C port
- 3x 3.6V 18650 Li-Ion batteries
- Optional solar panel
Dimensions – TBD
Temperature Range – 40°C to 85°C
Enclosure – Plastic ABS, IP64 rated

SolidRun provides a Yocto-based Linux distribution with libraries, BSP, and drivers. The AI software supports Keras, Tensorflow, PyTorch, and ONNX models with tools to compile and optimize the model before deployment on the device. There’s documentation on the product page, but it needs to be updated since the software section still points to NXP i.MX 8M Plus resources.

Target applications include border and perimeter surveillance, wildlife monitoring and conservation, environmental and weather monitoring (e.g., detect floods, landslides, and monitor air or water quality in remote locations), Smart Agriculture, and remote industrial equipment monitoring (e.g., predictive maintenance).

The product page has more information, but it is currently a mess. Besides the aforementioned software documentation issue, the page focuses on the Hailo-15 AI vision development platform, but the “order options” tab reads “Choose your SolidSense AIoT RZ/V2N Configuration” with various hardware options. Neither models are sold on the online store, so you’d need to request a quote to get pricing for your configuration

Jean-Luc Aufranc (CNXSoft)

Jean-Luc started CNX Software in 2010 as a part-time endeavor, before quitting his job as a software engineering manager, and starting to write daily news, and reviews full time later in 2011.