Asus has introduced the PE3000N, an AI platform leveraging Nvidia’s Jetson Thor architecture, incorporating the Nvidia Blackwell GPU and a 14-core Arm CPU. This system is crafted for the robotics sector, providing significant AI processing power of 2,070 FP4 petaflops while maintaining space and energy efficiency. It supports various environments with its robust design featuring industrial-grade connectors and a low-profile chassis, making it suitable for challenging outdoor settings.
The PE3000N facilitates high-bandwidth sensor fusion and advanced machine vision, supporting up to four 25GbE links and 16 GMSL cameras. Its extensive DC input range and operational temperature gamut allow flexibility for factory automation, autonomous vehicles, and smart city infrastructures. The system also boasts modular I/O layers for evolving sensor technologies, including PoE and CAN interfaces, while supporting reliable sensor synchronization and communications through LTE, 5G, and GNSS modules. Asus indicates that the PE3000N enables independent intelligent analytics by reducing reliance on cloud connectivity, facilitating physical AI applications within various industries.
👉 Pročitaj original: CIO Magazine
