PlanetSpark launches Xilinx-based AI edge computing box

PlanetSpark, a Singaporean technology company under the World Health Group that focuses on providing hardware solutions in the fields of artificial intelligence and the Internet of Things, officially launched the Edge AI Box X7, an AI edge computing box independently developed by its technical team . This product belongs to PlanetSpark's Black King Kong Edge AI series. It is an AI intelligent computing device based on Xilinx Zynq Ultrascale+MPSOC, designed for edge artificial intelligence and machine vision, and can be easily installed and debugged on site and deployed on the move. With its included middleware and support for many common peripherals and interfaces, this product can help developers quickly start industrial IoT applications, computer vision applications, and embedded vision applications. In addition, it can be paired with PlanetSpark's partner applications and perform real-time analysis.

Figure 1: Edge AI Box X7

Edge AI Box X7 has the following four advantages:

1. Powerful processing capability and ultra-low power consumption

With support for multiple high-speed interfaces such as USB3.0, Tri-mode Gigabit Ethernet and SATA 3.1, developers can use it to quickly start their embedded vision designs.

2. Rich interface and peripheral options

With its efficient, scalable and programmable IP core, it can be customized for many different applications to achieve a perfect combination of low power consumption and high performance.

3. Provide pre-trained models (provided by Xilinx)

A comprehensive set of optimized models is provided, which developers can use to find the closest model and start re-tuning it for their target application.

4. Provides preset partner AI programs that can be quickly put into production

Partnering with AI application providers such as Aupera, PlanetSpark provides commercial-grade AI applications such as ANPR and Crowd Statistics for real-time analysis. As a technology partner of Xilinx, PlanetSpark also provides a complete development environment and tool support.

Development Environment

Vitis™ AI development environment is Xilinx's development platform for AI reasoning on Xilinx hardware platforms (including edge devices and Alveo™ cards). It consists of optimized IP, tools, libraries, models, and example designs, designed with high efficiency and ease of use in mind, unleashing the full potential of AI accelerators on Xilinx FPGAs and ACAPs.

Open to all users are popular frameworks such as Pytorch, Tensorflow, Tensorflow2, Caffe, and ready-made deep learning models. AI Model Zoo provides optimized and reconfigurable AI models, enabling users to achieve faster application development, faster processing power, and more efficient productization on all Xilinx platforms.

Figure 2: User Application

Figure 3: AI Model Zoo

Technical specifications

Figure 4: Edge AI Box X7 interface

Figure 5: Technical specifications

Figure 6: Edge AI Box X7 is stackable

Application Scenario

Smart Parking

Utilize various forms of technology, including sensors and edge computing , to create a real-time heat map of parking spaces in a specific area. This information can be used in a variety of ways, including but not limited to: using applications to help drivers easily find nearby parking lots, redirecting traffic to prevent regional traffic congestion in the absence of parking spaces, and communicating this information to parking lot management in the case of illegal parking. Realize intelligent management and control, vehicle positioning and dispatching, vehicle remote detection and service, traffic status perception, etc.

People flow statistics

Use sensors and cameras to capture data on human traffic in a specific area. Collect relevant data, such as the movement of people, the number of people, and other forms of data. It can promote the coordinated and efficient operation of urban human traffic and improve the standardization, precision, and intelligence of urban construction and management. Make full use of information technology, deepen the intelligent construction of social security monitoring dynamic video systems and data mining and utilization, integrate human traffic data and social monitoring resources, and improve and deepen the "smart and safe city".

Figure 7: Edge AI Box X7 dimensions are 151x111x53mm  

In addition, it is worth mentioning that the Edge AI Box X7 has a small size of 151x111x53mm and is very light, which can greatly save installation space and facilitate mobile installation.