Aetina Corporation has announced mass-production availability of the DeviceEdge AIE-VX28/18-IP, a fanless, IP66-rated edge AI platform designed for heavy-duty vehicles, off-highway machinery, and roadside infrastructure. Powered by Nvidia Jetson AGX Orin, the system supports up to 12 simultaneous camera inputs, enabling 360-degree vision analytics for applications such as obstacle avoidance, pedestrian detection, and equipment safety. Designed for construction, agricultural, mining, and rail vehicles, the platform combines MIL-STD-810H durability with four power over Ethernet (PoE) camera connections and eight GMSL2 inputs. Its sealed enclosure protects against dust and water, while waterproof M12 connectors with threaded locking help maintain reliable connections under vibration. The system operates from -25°C to 70°C and accepts a 9-48 VDC input, with ignition power control for surge protection and safe shutdown. It also carries E-Mark (E24) automotive certification. The four M12 PoE ports provide a combined 60-W power budget for IP cameras, while eight GMSL2 cameras connect through mini-FAKRA connectors. This combination supports different camera configurations, with PoE simplifying cabling and GMSL2 providing low-latency video transmission over longer distances across large vehicle chassis. The Nvidia Jetson AGX Orin module delivers up to 275 TOPS of AI performance for concurrent deep neural networks and real-time vision processing using Nvidia JetPack 6.2 or later. With JetPack 7.2 and its Super Mode, the Jetson AGX Orin 32-GB module increases AI performance from 200 to 241 TOPS, providing additional processing capacity without requiring a higher-cost module. Connectivity includes M12 10 GbE, isolated CAN FD, and RS-232/422/485 interfaces. M.2 expansion supports NVMe storage, Wi-Fi, and Bluetooth, as well as RTK positioning and LTE/5G connectivity modules. Target applications include blind-spot monitoring and geofence alerts for construction machinery, autonomous agricultural equipment, outdoor autonomous mobile robots, and roadside traffic monitoring. At Smart City Expo World Congress 2026 in Barcelona, Aetina is demonstrating a loading-dock safety application using eight concurrent GMSL2 camera streams and the Nvidia DeepStream SDK to identify workers near a truck. The same approach can be applied to forklifts, excavators, and other vehicles with restricted visibility. The AIE-VX28/18-IP is available now for mass-production orders. Aetina also provides an optimized board support package, long-term technical support, and customization services covering chassis design and system validation. Validated peripherals include Innodisk automotive GMSL2 cameras, industrial storage, and 4G/5G connectivity modules. Related articles Aetina Introduces Palm-Sized In-Vehicle Edge AI Platforms Aetina Launches Edge AI Systems for Robotics The post Aetina Unveils Rugged Edge AI System for Heavy-Duty Vehicles appeared first on Embedded .