Taiwanese firm Asus IoT has unveiled a compact edge-AI platform for robotics and intelligent automation.

The PE3000N is built round the Nvidia Jetson Thor platform, with Nvidia Blackwell GPU, a 14-core Arm CPU and 128Gbyte of LPDDR5X memory, enabling 2070 FP4 TFlops of AI processing power in a space-efficient form factor.

This makes it suitable for integration into robotic systems where space and energy efficiency are critical. With its robust architecture, the Jetson T5000 module helps developers and integrators achieve new levels of autonomy, sensor fusion and AI-driven control for industrial, commercial and smart infrastructure deployments. 

Engineered for durability, it incorporates MIL810H industrial-grade connectors and a low-profile chassis to withstand demanding operating conditions. With support for up to four optional 25GbE links and 16 GMSL cameras, it enables high-bandwidth sensor fusion and machine vision, even in difficult environments.

The 12 to 60V DC input and ignition support provide stable, battery-friendly operation across diverse settings from factory floors and autonomous vehicles to smart-city infrastructure.

With an operating temperature from -20 to +60˚C, it ensures resilient performance and secure data handling, making it suitable for mission-critical robotics, automation and edge AI deployments. 

Versatility is provided by the scalable chassis and modular IO layers that support interfaces such as PoE, GMSL, Can and QSFP28 to accommodate evolving sensor and network needs. The optional second stack enables vertical expansion within a compact 2U height, allowing for project-specific IO customisation.

Additional capabilities include PTP/PPS for precise sensor synchronisation, onboard TPM 2.0 for security, and support for LTE, 5G and GNSS modules, broadening deployment options across industry sectors.

The computational power supports efficient execution of generative AI models, such as visual language models (VLMs) and large language models (LLMs), as well as real-time video analytics and autonomous control tasks. Also, to deliver a seamless cloud-to-edge experience, it runs Nvidia AI technologies or physical AI applications, including Nvidia Isaac for robotics and Holoscan for real-time sensor processing, and even building and deploying video analytics AI agents using Nvidia agentic AI workflows such as Blueprint for Video Search & Summarization (VSS) from Metropolis (www.nvidia.com/en-gb/autonomous-machines/intelligent-video-analytics-platform/).

By processing vast amounts of data and facilitating rapid decision-making directly at the edge, it ensures autonomous operations and intelligent analytics with low latency, eliminating reliance on cloud connectivity and increasing responsiveness for modern robotics and industrial automation.

For more information, visit www.asus.com/networking-iot-servers/aiot-industrial-solutions/all-series/filter/.