Advantech (2395.TW) is doubling down on edge AI innovation through a new collaboration with Qualcomm Technologies, Inc. and Edge Impulse, a Qualcomm company. The goal: to accelerate the deployment of next-generation edge AI solutions powered by the Qualcomm Dragonwing™ IQ-9075 processor.
The partnership combines Advantech’s edge AI hardware, Qualcomm’s powerful Dragonwing chip, and Edge Impulse’s developer ecosystem, creating a full-stack environment for building, training, and deploying AI models at the edge with industrial-grade reliability.
The Dragonwing IQ-9075 powers Advantech’s new AIR-055 and AFE-A503 platforms, delivering up to 100 TOPS (trillion operations per second) of AI performance through dual Qualcomm® Hexagon™ NPUs and an 8-core Qualcomm® Kryo™ Gen 6 CPU. Built for rugged use, the chip also promises 10+ years of product longevity, catering to long-term deployments in industries like smart manufacturing, robotics, and transportation.
“By combining Qualcomm Technologies’ powerful Dragonwing IQ-9075 processor and Edge Impulse’s proven platform, we are creating new AI-ready solutions with a strong foundation for intelligent edge applications,” said Miller Chang, President of Advantech’s Embedded IoT Group. “Together, we’re enabling enterprises to unlock new levels of performance, scalability, and efficiency.”
Qualcomm’s Dragonwing Brings Industrial Muscle to Edge AI
This partnership deepens Qualcomm’s presence in the industrial and embedded AI space—long a focus area for the chipmaker’s expansion beyond mobile.
“Our expanded collaboration with Advantech marks a pivotal step in bringing high-performance edge AI to industrial and embedded applications,” said Nakul Duggal, Group GM for Automotive, Industrial & Embedded IoT at Qualcomm Technologies. “This will empower industries to harness the full potential of intelligent, autonomous systems.”
The addition of Edge Impulse strengthens the ecosystem by making AI model training and deployment more accessible to developers and engineers—bridging the often-complex gap between data science and edge hardware integration.
AIR-055: Compact Edge AI System for Vision AI and VLM/LLM
Proofread version:
Powered by the Dragonwing™ IQ-9075 processor with up to 100 TOPS of AI performance, the Advantech AIR-055 edge system offers a rich set of connectivity options, including USB, RJ45, HDMI, and M.2 interfaces. It incorporates MIPI-C technology co-developed with Appropho and Orbbec, enabling direct MIPI-CSI connections through a USB-C cable to 2D or 3D industrial cameras—no bridge ICs required. Supporting 4K resolution and 200 cm transmission distance, the AIR-055 is optimized for smart manufacturing, intelligent transportation systems (ITS), retail, public safety, robotics, and autonomous mobile robots (AMRs).
AFE-A503: Robotic Controller for Next-Gen AMRs
Proofread version:
Built around the Dragonwing IQ-9075 processor, the Advantech AFE-A503 is purpose-designed for outdoor AMRs and delivery robots. It features four Ethernet ports for 2D/3D LiDARs and PoE cameras, eight GMSL interfaces for RGB and 3D camera combinations, and isolated CANFD and RS232 connections for IMU integration. Its 20–36V power design supports all-day operation on battery-powered systems.
The AFE-A503 also includes the Advantech Robotic Suite, preloaded with ROS2, AI acceleration SDKs, and sensor node integration—enabling developers to achieve faster time-to-market for robotic and automation applications.
Why It Matters: The Edge AI Arms Race Heats Up
This collaboration highlights an industry-wide trend: edge AI hardware is getting smaller, smarter, and more power-efficient—while becoming easier to program and deploy thanks to platforms like Edge Impulse.
With Qualcomm’s Dragonwing IQ-9075 providing the horsepower, Advantech’s industrial reliability anchoring the hardware, and Edge Impulse simplifying the developer experience, this trio is building what could become a benchmark ecosystem for AI-driven edge computing.
Expect to see early use cases in smart factories, logistics automation, and robotics, where AI at the edge is quickly moving from prototype to production.
Power Tomorrow’s Intelligence — Build It with TechEdgeAI
