Topic : New products and services
New Taipei City, Taiwan, September 21, 2022 /PRNewswire/ — Etinean EdgeAI solutions provider, presented its end-to-end AI management solution this week as part of a session on request from NVIDIAGTC. This solution will help Aetina’s partners and customers successfully adopt EdgeAI using NVIDIA’s AI development and deployment tools, as well as Aetina’s NVIDIA training and inference platforms, powered by the ‘IA.
During the session, Jeremy Pan, Product Manager of AI solutions at Aetina, immersed himself in Aetina’s solution and shared a powerful application case of automatic optical inspection (AOI) powered by AI in the factory of a Yilan customer, Taiwan.
The Challenges for EdgeAI Adoption
To adopt EdgeAI, system integrators and developers must train AI models and deploy them to Edge devices. The process of training an AI model involves collecting and labeling large amounts of data using high-performance computing platforms, which can lead to high training costs. These factors make the training process difficult and time-consuming for system integrators and developers.
Deploying an AI model can also be challenging when system integrators and developers manage multiple remote Edges and multiple locations. Challenges of deploying AI models include optimizing inference models to run efficiently on Edge devices, lengthy deployment processes, difficulties in team communication and collaboration, high tracking costs caused by a large number of AI models, security issues and high costs.
An end-to-end AI management solution
Aetina provides an end-to-end solution to help its partners and customers develop and deploy AI applications for the Edge. This solution consists of the Edge platforms NVIDIA Certified from Aetina and AI model development and deployment tools from NVIDIA. These tools include NVIDIAFleetCommandtm and the NVIDIAAIEnterprise software suitewho provides corporate support for the toolkit NVIDIATAO as well as the NVIDIATriton InferenceServert™.
With Aetina’s NVIDIA-Certified Edge Platforms and other NVIDIA tools, Aetina streamlines the AI training and deployment process by helping partners and customers adopt EdgeAI faster.
An application case in a smart factory
One of Aetina’s customers, a global provider of industrial embedded flash memory and random access memory (DRAM) solutions, has planned to add AI-powered AOI systems to its factory for improved productivity. Aetina’s customer-produced flash and DRAM products are small and complex electronic components designed for harsh environments and applications. The producer of these components needed an AOI system capable of handling high resolution image recognition with high processing speed. Aetina helped their client develop a prototype AI-powered AOI system.
A team at Aetina first trained an AI model capable of finding faulty products in factory production lines for AOI system with NVIDIATAO on Aetina’s AI training platform, SuperEdge AIP-D422, then uploaded the model to the NVIDIAGPUCloud (NGC). Using NVIDIAFleetCommandtm, the team remotely deployed the model to Aetina’s AI inference platform, MegaEdge AIP-FQ47, in Aetina customer’s factory from the NGC, thus successfully developing the prototype AOI system.
With AOI cameras, the prototype system is now able to dramatically speed up the inspection task of spotting misplaced corporate stickers on DRAM products with high accuracy in less than a second, compared to a human inspector which takes about ten seconds to perform the same task. In the future, when the AI-powered AOI system is fully built, it will be installed in Aetina’s customer’s factory to perform inspection tasks on multiple production lines.
Additionally, NVIDIAFleetCommandtm enables the Aetina team to quickly update the AI model on Edge devices in Aetina’s customer’s factory production lines, enabling efficient remote AI management. This is the key to a successful EdgeAI deployment.
The end-to-end AI management solution is part of Aetina’s Pro-AI service, which helps global partners and customers adopt AI for different vertical applications in addition to AOI in factories, through to Aetina’s state-of-the-art AI hardware and software.
These press releases may also interest you
Press release sent on September 21, 2022 at 3:00 p.m. and distributed by: