The AIR-055 addresses this demand by combining 100 TOPS of AI performance, long-range camera connectivity, over 20 I/O interfaces and expansion slots, and integrated software tools into a compact device to accelerate the adoption of multimodal AI in smart manufacturing, robotics, automated optical inspection, intelligent transportation, and smart retail applications. By integrating Advantech WEDA (WISE-Edge Developer Architecture), Edge AI SDK, BSP Launcher, and Qualcomm AI Hub resources, the AIR-055 lowers application development barriers and shortens the process from proof of concept to large-scale deployment.
Based on the Qualcomm Dragonwing IQ9, the AIR-055 delivers up to 100 TOPS of AI performance and 36 GB of LPDDR5 memory, resulting in 1.5 to 4.5 times greater memory efficiency across various AI models. With an ultra-low power consumption of 60W, it accelerates generative AI, machine vision, and multi-model inference at the edge. For AI vision applications, the system also supports USB and PoE industrial cameras and can be integrated with GMSL modules for longer-range deployments or in harsh environments. This makes the AIR-055 an ideal solution for automated optical inspection (AOI), autonomous mobile robot (AMR) perception, robotic vision, and smart manufacturing applications that require low-latency image capture and real-time AI processing. The AIR-055 also features over 20 I/O interfaces and expansion options, including USB, LAN, COM, DIO, CAN, video outputs, M.2 NVMe, Wi-Fi 6E/Bluetooth, 5G LTE, and PoE expansion. Its compact, fanless design, measuring 190 × 174 × 57.8 mm, supports operating temperatures from -20 to 55 °C, making it ideal for uninterrupted industrial deployments in space-constrained environments.
A software platform ready to streamline AI deployment and management at the edge:
To simplify AI development and large-scale deployment, the AIR-055 integrates Advantech software tools, such as BSP Launcher, Edge AI SDK, and Advantech WEDA. BSP Launcher enables remote BSP updates, reducing on-site maintenance. Edge AI SDK provides pre-validated AI frameworks and runtime environments, and supports over 300 optimized models from Qualcomm AI Hub to expedite model selection, integration, and validation. With WEDA, the AIR-055 also enables remote deployment, OTA updates, and device management, helping businesses reduce the complexity of managing AI systems at the edge on a large scale.
Aetina launches AIE-KT78 and AIE-KT68 with NVIDIA Jetson Thor for robotics and AI at the edge
Aetina Corporation has announced the availability of its new edge AI systems, the DeviceEdge AIE-KT78 and AIE-KT68, based on NVIDIA Jetson Thor. Both devices utilize the NVIDIA Blackwell architecture and combine AI processing capabilities, connectivity for...
