Nvidia Corp. is making it easier to deploy artificial intelligence (AI) at the edge in the harshest environments with the introduction of the Jetson AGX Xavier Industrial module. Aimed at the toughest safety-critical environments, the AI embedded industrial module can be used in applications for intelligent video analytics, optical inspection, robotics, and computer vision, across harsh environment sectors. These include industrial, aerospace, defense, construction, agriculture, logistics, inventory management, delivery, inspection, and health care.
“The powerful Nvidia Jetson AGX Xavier platform has set the standard for robotics and automation in diverse industries, providing real benefits to customers leveraging AI and deep learning in their operations,” said Murali Gopalakrishna, head of product management, autonomous machines and general manager for robotics at Nvidia. “Now with the Jetson AGX Xavier Industrial module we extend those benefits with ruggedized features and enhanced safety for harsh environments in industries like manufacturing, oil & gas, construction, and aerospace.”
The industrial system-on-module retains a compact size, efficiency, and performance with extended industrial ranges for shock and vibration, temperature, and lifetime. It offers the same key accelerators and high-speed I/Os included with Jetson AGX Xavier, and includes a 512-core Nvidia Volta GPU with 64 Tensor Cores, two NVIDIA deep learning accelerators, two vision accelerators, an eight-core Nvidia Carmel Arm CPU, and an encoder and decoder, delivering up to 30 trillion operations per second (TOPS) of AI performance.
The Jetson AGX Xavier Industrial adds new functional safety capabilities and are built with components tested to industrial standards. This enables the module to withstand severe shock and vibration and operate over extended temperature ranges. As an example, the module can be used in the oil & gas industry to provide real-time insights based on monitoring or inspection of pipelines, valves, equipment, and maintenance work.
The new Safety Cluster Engine (SCE) contains dual Arm Cortex-R5 processors that can be used for integrated fault-detection mechanisms and lock-step subsystems and enable built-in-system test. The Cortex-R5, which is in an always-on domain, can be used for safety and error correction functions, said Nvidia. The functional safety capabilities are supervised by the SCE, making it suitable for safety-certified industrial-grade products.
The module also adds new features such as error correction codes, single error correction, double error detection, and parity protection to deliver internal RAM resilience, address and data bus error detection and correction, and IP resiliency in industrial applications, which are required to deploy AI in extreme environments, said the company.
Security features include hardware-verified secure boot, hardware-accelerated cryptography, support for encrypted storage and memory to protect customer software.
The Jetson AGX Xavier Industrial, with the Nvidia CUDA-X accelerated computing stack and JetPack SDK support, is a software-defined platform enabling cloud-native capabilities. The CUDA-X acceleration, free production-ready pretrained models on NGC and NVIDIA Transfer Learning Toolkit help developers build and deploy deep learning and AI training and inference systems.
The AI embedded module is pin-, software- and form-factor compatible with the existing Jetson AGX Xavier module, which makes it easier to upgrade. The Jetson AGX Xavier Industrial module and developer kit is available to order now and will be available in late July.