Advantech’s VEGA-300 series AI acceleration modules are built for AI at the Edge. They provide highly efficient add-on modules in different form factors including M.2, mini PCIe, and PCIex4. They can be easily integrated into existing systems. The integration is seamless and will simultaneously reduce CPU workload. The Edge AI Suite software allows users to quickly enable AI inference on edge devices. That allows the realization of AI and vision analytics applications in many industries including retail, transportation, medicine, and manufacturing.
A Use Case Story
AI-power smart street lights
There are thousands of street lights in cities throughout the world. They help drivers and pedestrians find their way to their destinations safely. When equipped with AI technology, street lights can be used to not only light up roads but also support city services such as air quality, temperature monitoring, humidity, and also parking and traffic management.
The solution is the M.2 AI Modules: Vega 320
The city planners had already implemented the edge computing system, to serve as a gateway computer for collecting data, including humidity and temperature, which helped facilitate operational analysis. That system was updated to the VEGA-320 edge AI acceleration modules, embedded into cameras for analyzing multi-streams of video data. The system also performed real-time tasks such as finding available parking spaces. The results were encouraging with the city seeing a 10% to 15% increase in the utilization of roadside parking.
The key specifications of the VEGA-300 series include
- VEGA-320
- 1x Myriad X MA2485 SoC
- M.2 2230 (Key A + E) form factor
- PCIex1, USB 2.0
- Power 3.8W
- 22 x 30 x 3.63 mm
- -20° ~ 60°C
- Driver support
- Windows 10 Enterprise
- Ubuntu 16.04.3 LTS(64-bit)
- CentOS 7.4(64-bit)
- VEGA-330
- ½ x Myriad X MA2485 SoC
- Full-size miniPCIe
- USB2.0, PCIe x1
- Power 3.8W or 7.6W
- 30 x 50.95 x 4.86 mm
- -20° ~ 55°C
- Driver support
- Same as VEGA-320
- VEGA-340
- 4 and 8 x Myriad X MA2485
- PCIe x4 Low profile
- PCIe Generation 2
- Power 16.8W or 28W
- 171.1 x 68.9 mm
- -20° ~ 60°C
- Driver support
- Ubuntu 18.04,
- LTS(64-bit)
- Windows 10 Enterprise
Built-in Edge Suite of software tools
Advantech has developed a suite of tools they call the Edge AI Suite, for accelerated deep learning inference on edge devices. The systems are integrated with Intel’s Distribution of OpenVINO toolkit. The Edge AI Suite provides a deep-learning model optimizer, pre-trained models, a user-friendly GUI, and a model optimizer.
OpenVINO Toolkit
- OpenVINO R3.1 Integrated
- Performance boosted inference
- Heterogeneous executions across CPU/VPU/GPU
Pre-trained models
— ADVERTISMENT—
—Advertise Here—
- Rapid object detection (YOLOv3)
- Human pose estimation
- Facial recognition
3rd Party AI SDK
- License plate recognition
- Vehicle classifications
- Facial recognition
Deployment Wizard
- CPU/VPU/ Memory monitoring
- Supports deep learning frameworks in various circumstances
- Quick start tutorial
For more information on Advantech’s VEGA-300 Series modules, see the product page here.
Advantech | advantech.com
Sponsor this Article
For the past 8 years, I have been writing about embedded technologies, added to my technical, academic, and medical editorial experience, with companies like Elsevier and Cambridge University Press. I tell people to read what I write, not try to pronounce my last name. I am always available for comments and suggestions you can reach me at product-editor@circuitcellar.com and I promise I will take the time to reach back out to you. I live in the North East with my wonderful family.
