The Brand-New Interaction Between Clients and Environment
Since the technology of edge computing has became more and more mature, many AIoT applications have started testing in public places. ITRI released one of their applications, which is combined with 5G WiFi Fog Computing and Augmented Reality, and began the live demo at several station in Taipei.
In this case, developers and system integrators build the system by Aetina Jetson Platform as the computation central, then connect it with 5G WiFi AP and let the clients link the 5G WiFi fog computing LAN with their mobile phone for device-localization. Instead of computing in end-devices and insufficient memory of the mobile phone, they choose fog computing to offload the computation and reduce the power consumption of clients’ mobile phone. With the APP, the client is allowed to connect the system and sent most of work to the 5G fog computing LAN, then do the real-time recognition and localization. After the positioning, they can start to enjoy the convenience brought by the AR navigation, and at the same time, to obtain the shop information and related discount coupons through using the phone camera focus on the sign board. It’s much more convenience to find the shop you looking for and get the information of it. This system successfully lowers the cost of Off-line advertisement and change the interaction between clients and public place. The store owner can achieve the purpose of Online-to-Offline (O2O) by using this brand-new AR navigation and advertisement platform.
The main difference between fog and edge computing is distributed computing. Due to the huge crowds in the shopping mall, several fog nodes are set up in the indoor space and required to compute the data as much as possible. That’s why GPGPU plays an important role in this case, every fog node can do the parallel-computing separately and share the result through WiFi or Internet.
In the other hand, the reasons to choose Aetina Jetson Platform is Small-Form-Factor, high quality edge computing performance and Support Multiple Driver-Ready I/O Extension Module. We provide low-power envelope AI applications in ultra-small form factor, and other size for different purposes. In the same time, Aetina Jetson Platform has qualified for VID and I/O extension module, and we also co-work with ITRI which able to provide software developing and integrating. Trough those benefits developers can save the time and cost on system and peripheral integrating, it’s suitable for every embedded AIoT applications, such as smart retail, smart surveillance, smart factory, robotic and intelligent logistics.
Computing Module | ||
---|---|---|
Nvidia AGX Xavier | Nvidia Jetson TX2 | Nvidia Jetson TX1 |
512 Tensor cores | 256 CUDA cores | 256 CUDA cores |
2.5 TFLOPS | 1.5 TFLOPS | 1 TFLOPS |
16 GB 256-Bit LPDDR4x memory | 8 GB 128-Bit LPDDR4 memory | 4 GB 64-Bit LPDDR4 memory |
More Detail | More Detail | More Detail |
Carrier Board | |||
---|---|---|---|
For AGX Xavier | For Jetson TX2/TX1 | ||
NSO-MI-Xavier | ACE-N310-A | ACE-N622 | ACE-N510 |
Upcoming product AX-710-A Support 10G LAN |
Support 6x MIPI Camera Module |
Extension I/O & Nano-ITX | Credit-Card size |
More Detail | More Detail | More Detail | More Detail |