ASUS Launches Rugged PC for Edge AI

- The Intel Core Ultra processors (either the 265U or 235U model) inside the PE1000U feature a hybrid architecture that combines a traditional CPU, an integrated GPU, and a Neural Processing Unit (NPU) for AI acceleration. This allows it to simultaneously handle real-time motion control, run AI inference models, and process graphics workloads. - For robotics-specific interfaces, the device includes two CAN bus ports for deterministic communication with vehicle or robotic arm controllers, up to four Ethernet ports for connecting sensors like cameras and LiDAR, and an 8-bit isolated digital I/O module for triggering industrial components. - The unit is built to the MIL-STD-810H US military standard for shock and vibration (withstanding 5 Grms), has an IP40 dust-proof rating, and operates in temperatures ranging from -25°C to 70°C, suiting it for factory floors or semi-outdoor use. - The system's onboard AI capabilities are rated for up to 22.5 TOPS (trillion operations per second), providing a specific performance metric for on-device machine vision and AI inference tasks. - Its DIN rail mount is a key feature for industrial automation, as it's a standardized metal rail system used inside control cabinets, allowing the PC to be easily integrated alongside other common hardware like PLCs (Programmable Logic Controllers). - Wireless connectivity for mobile robots is enabled through M.2 expansion slots, which can be fitted with modules for Wi-Fi 6E and cellular 5G/LTE communication for fleet management and data transfer. - While ASUS is a major player, other companies like Premio, Portwell, and Winmate also produce rugged, fanless industrial PCs for edge AI, indicating a competitive market for this type of hardware. - This x86-based architecture competes with other popular edge AI platforms in robotics, most notably NVIDIA's Jetson series, which is built on ARM architecture and is also widely used for AI-powered autonomous machines.