The most powerful Embedded mini-PC for 3D perception and AI computing.
Powered by NVIDIA® Jetson™ Xavier, the ZED Box is an industrial AI gateway for challenging environments. It delivers the required GPU horsepower to run modern neural networks and process 3D sensor data in real time, allowing for a complete, self-managed AI system.
The ZED Box has built-in NVIDIA® Jetson Xavier GPU, for familiar development and fast 3D and machine learning computation at the edge.
Easily and seamlessly ingest and process live streams of ZED cameras, augment the data and transmit it by Ethernet or Wifi.
The ZED Box can be used for a growing number of industrial use-cases such as edge analytics, autonomous robotics, interactive experiences, and many more.
With a rugged aluminium and fanless chassis, this embedded computer packs the power of a supercomputer into a small form factor (109 x 92 x 52 mm).
Deploy edge devices anywhere in the field using PoE+, enabling long-range and outdoor coverage, simplified cabling infrastructure and easier maintenance.
The Edge AI Gateway is a ruggedized and field-tested embedded system, which makes it suitable for 24/7 deployment in many types of environments.
Built for edge computing, the Stereolabs ZED Box can be mounted in a robot, on a wall or DIN rail at the edge of the network enabling you to collect, secure, analyze and act on data from any camera and sensor. Powered by an NVIDIA Jetson processor, run AI locally and send only meaningful data over the network.
Connect your camera and set up your gateway in minutes. Once installed, deploy your app and start processing 3D video feeds with CV and AI in real-time.
Mount the ZED Box on a wall or in a mobile system. Connect your 3D camera and power source, or use PoE+ for simplified cabling infrastructure.
Deploy and run your application on the ZED Box to start processing 3D data in real time at the edge.
Use a modern, cloud-based platform to monitor your live video stream, remotely control your gateway and sensors, deploy apps and collect data.
Combine ZED stereo cameras with a ZED Box powered by NVIDIA Jetson to bring unprecedented intelligence to your application.
NVIDIA® Jetson™ TX2 module gives you exceptional speed and power-efficiency in an cost-effective embedded AI computing device.
NVIDIA® Xavier NX delivers up to 21 TFLOPS, making it ideal for high-performance compute and AI in embedded and edge systems. Xavier NX is the platform of choice to run multiple modern neural networks in parallel and process high-resolution 3D data.
Stereolabs ZED Box comes with two NVIDIA® Jetson™ Module options. Get an NVIDIA TX2 NX or NVIDIA Xavier NX for more performance.
| ZED BOX - TX2 NX | ZED BOX - Xavier NX | |
|---|---|---|
| Jetson Production Module | TX2 NX | Xavier NX |
| Carrier | Custom Low-Profile Optimized NVIDIA Jetson Carrier Board | Custom Low-Profile Optimized NVIDIA Jetson Carrier Board |
| GPU | 256-core NVIDIA Pascal™ GPU | 384-core NVIDIA Volta™ GPU with 48 Tensor Cores |
| CPU | Dual-Core NVIDIA Denver 2 64-Bit CPU and Quad-Core ARM Cortex-A57 MPCore | 6-core NVIDIA Carmel ARM 64-bit CPU |
| DL Accelerator | - | 2x NVDLA |
| Memory | 4GB LPDDR4 - 51.2 GB/s |
8GB LPDDR4x - 59.7 GB/s |
| Jetson Power Modes | 7.5W | 15W | 10W | 15W |
| Storage | eMMC 16GB SSD 256GB M.2 NVMe |
|
| I/Os | 1x USB 3.0 1x USB 2.0 OTG (Flash) 1x Gigabit Ethernet 1x HDMI |
|
| Operating System | Ubuntu 18.04 (L4T) | |
| Power | PoE+ (30W) or DC-IN (12-48V). | |
| Dimensions | 109 x 92 x 52 mm (LxWxH) 4.3 x 3.6 x 2.0 inches |
|
| Weight | 1 kg (2.20lb) | |
| Operating Temperature | -20°C to +60°C (-4°F to 140°F) with 0.5 m/s airflow | |
| Mounting Options | 84x45mm mounting holes, wall (brackets provided) and DIN-rail mounting | |
| Case | Industrial black aluminium case passively cooled | |
Hardware + Software
Benefit from leading-edge multi-sensor, high resolution smart stereo cameras ready for deployment in the field.
Well-documented and easy-to-use C++ and Python API for building real-time space-aware applications on desktop and embedded platforms.
Build with ROS, OpenCV, deep learning frameworks and many more tools and plugins compatible with ZED.
Use a modern, cloud-based platform to monitor remotely your cameras and collect data.
