Industrial Edge AI Computer Leveraging NVIDIA Jetson Platform
Anewtech is expanding its range of edge AI systems built on the Jetson platform for smart city, intelligent manufacturing, and AIoT applications.
The compact, energy-efficient MIC AI edge inference computer leverages the powerful new NVIDIA® Jetson platform, is designed to withstand shock and vibrations in industrial environments with temperatures ranging from -10°C to 60°C, making it ideal for ruggedized compute-intensive embedded AI applications at the edge.
MIC AI Inference series is a passively cooled system design embedded with the full line up of Jetson SOMs—including the new Jetson Xavier NX, Jetson AGX Xavier, Jetson Nano, and Jetson TX2. These systems enable AI applications in harsh environments—like road-side deployment or on manufacturing production lines. AI IPC supports flexible industrial I/O interfaces for data acquisition and communication, such as installing add-on cards like frame grabbers and wired/wireless modules to meet different application needs.
Edge AI inference system powered by the Jetson platform supports flexible industrial I/O and remote device management to facilitate AI deployment at the edge for various applications. The new Jetson Xavier NX with 384 CUDA® cores, 48 Tensor Cores, and cloud-native capability is the ideal platform to accelerate AI applications. Bundled with the latest NVIDIA Jetpack™ 4.4 SDK and smaller than a credit card (70x45mm), the energy-efficient module delivers server-class performance — up to 21 TOPS at 15W, or 14 TOPS at 10W. This opens the door for embedded edge-computing devices that demand increased performance to support AI workloads but are constrained by size, weight, power budget, or cost.
Edge AI System provides flexible and modularized industrial I/O support for requirements in different vertical applications. Customers are able to do simple customization without changing system.
Application: AI in Transportation
Die Inspection for chip probing Printer line balance optimization Self-adaptive traffic signal control system Traditional traffic detection methods like microwave radar are costly and lack the necessary recording detail. By using AI, essential traffic metadata can be collected and integrated into a carefully designed dataflow for further analysis.
AD-MIC-720AI, adopted at roadside, grabs video streams from IP cameras through PoE and sends the analysis and inference results back to the server. The control server manages all the traffic lights automatically based
Application: Automatic AI IOT
Die Inspection for chip probing Die surface were traditionally inspected by human eyes under a microscope. To increase inspection efficiency and accuracy, AI is deployed in die processing equipment for surface defect inspection.
AD-MIC-770 controls the chip probe and camera. After image processing, MIC-770 passes the image to AD-MIC-730AI, where a trained AI model quickly provides image inference results for each die to MIC-770. The qualified dies can proceed to the die tray for IC packaging and final test(FT) processes.
Industrial AI IoT Solution Powered by NVIDIA®
As artificial intelligence (AI) becomes increasingly prevalent across a range of applications, the AI product series with the NVIDIA® is taking huge steps forward in making AI a reality for manufacturing, transportation, and smart city applications.