Edge AI solutions suite to support increased demand
Provider of intelligent IoT systems and embedded platforms, Advantech, has revealed the hardware and software integrated edge AI solutions powered by Intel Vision Accelerator Design with Intel Movidius Myriad X VPU and Intel Distribution of OpenVINO toolkit. According to ABI Research, shipments of devices with edge AI capabilities will grow to 1.2 billion units by 2023.
AI at the edge enables real-time decision making without the limitation of latency, cost, bandwidth, power consumption, and other considerations. Advantech offers integrated solutions for edge AI applications ranging from VEGA-300 series AI acceleration modules, inference systems, to ESRP series domain focused solution packages.
All of the solutions come with Advantech Edge AI Suite which combines Intel Distribution of OpenVINO toolkit, pre-trained models, and edge management with easy-to-use UI to accelerate AI inference at the edge.
“Advantech is working closely with Intel to apply artificial intelligence to edge computing,” said Miller Chang, President of Embedded IoT, Advantech. “Combined with Intel Movidius Myriad X VPU, and Intel Distribution of OpenVINO toolkit, our edge AI solutions help customers to rapidly develop and deploy vision-based AI applications such as robotic/AGV/drone, traffic monitoring, and medical Image analysis,” Chang said.
Advantech Edge AI Suite is a software development kit for edge AI inference acceleration. With Intel Distribution of OpenVINO toolkit integrated, Edge AI Suite provides a deep learning model optimiser, an inference engine, pre-trained models, and remote monitoring with a friendly GUI.
Customers can use Python-based API to quickly import their existing Caffe, TensorFlow, Apache MXNet, and ONNX trained models and deploy them to edge devices. Edge AI Suite also integrates VPU/CPU/Memory real-time status monitoring and model update functions for managing distributed edge devices.
Advantech Edge AI modules provide compact, low power plug-in modules in two form factors: VEGA-320 series m.2 modules with one Intel Movidius Myriad X VPU onboard, and VEGA-330 series miniPCIe modules with two Intel Movidius Myriad X VPUs onboard. Together they help accelerate computer vision and deep learning on your edge systems for a wide range of AI-based vertical applications.
With its software partner alignment, Advantech ESRP-LPR series provides an integrated solution for smart traffic applications. The solution includes an inference system, pre-trained models such as license plate recognition, vehicle classification, vehicle counting, and pedestrian detection. Plus, edge management software for fast development in AI-based smart traffic applications is included to help speed your application deployment.
VEGA-320 and VEGA-330 will be ready in April while ESRP-LPR will be ready in May.