March 18, 2026
news
Advantech will be attending NVIDIA GTC 2026 from March 16th to 19th in San Jose, California. The company will be exhibiting in booth #1134 and conference rooms #6077 and #6078 to highlight and discuss edge AI innovations powered by NVIDIA Jetson Thor and NVIDIA IGX Thor. Exhibit features include how physical AI and edge AI support real-world deployments across robotics, medical systems, smart logistics, and intelligent retail environments.
Ween Niu, general manager of Advantech North America, said that as AI rapidly evolves, the industry is moving beyond cloud-based model training to real-world physical applications, often referred to as physical AI. Advantech has a long-standing focus on edge computing and edge AI, building industrial-grade platforms designed for high-performance, reliable, and scalable deployments.
During the show, Advantech will demonstrate the ASR-A702 / AFE-A702, a robotics-enabled edge AI platform powered by NVIDIA Jetson Thor. Both work with Advantech Robotic Suite and NVIDIA Isaac ROS to provide recognition capabilities such as object detection, distance estimation, pose tracking, and VSLAM. This solution is designed for rapid integration of sensors and multi-camera systems.
Also on display will be the MIC-742, a platform designed for humanoid robot applications. Built-in NVIDIA Jetson Thor delivers up to 2,070 TFLOPS (FP4) of AI performance. The system works with the NVIDIA Holoscan sensor bridge to enable ultra-low-latency sensor-to-inference workflows for transformers and vision language action models.
For healthcare applications, Advantech will be showcasing the NVIDIA Jetson Thor-powered AIMB-294 medical AI board, designed to perform time-critical surgical instrument anomaly detection, organ segmentation, and AR overlays while operating at just 130 W without an additional GPU module. Combined with next-generation medical AI frameworks and Advantech’s edge AI software tools, this solution supports low-latency imaging pipelines and streamlined AI model deployment.
A further highlight is the medical-grade USM-500 platform built on NVIDIA IGX. The system facilitates time-sensitive multimodal sensor fusion and edge AI processing for AI-assisted surgery, intraoperative imaging, endoscopic video analysis, and robotic surgical guidance.
Introducing the AIR-075 visual AI platform, powered by NVIDIA Jetson Thor, developed for visual AI agents in intelligent logistics and warehouse management applications. Provide safety monitoring and incident response with 10GbE connectivity that supports high-bandwidth video streaming.
Advantech’s MIC-743 (built on NVIDIA Jetson Thor) facilitates real-time vision analytics, natural language queries, and rapid video summarization to enhance operational visibility, efficiency, and decision-making in logistics.
The DS-015 Edge AI system leverages NVIDIA Jetson Orin to run generative AI models through the Advantech Edge AI SDK. The system supports on-device large-scale language model (LLM) interactions without the need for cloud connectivity. Ideal for kiosks, smart retail, digital signage, and industrial edge deployments.
For more information, please visit advantech.com.
Chad Cox. The Embedded Computing Design Production Editor’s responsibilities include handling the news cycle, newsletters, social media, and advertising. Chad graduated from the University of Cincinnati with a BA in Cultural and Analytical Literature.
Other works by Chad
