Advantech announced the expansion of its Edge AI portfolio with the upcoming release of the MIC-735 powered by NVIDIA IGX T5000 module. Built on NVIDIA IGX Thor platform, this industrial-grade system is designed to bring real-time physical AI directly to the Edge by combining high-speed sensor processing, enterprise-grade reliability, and functional safety.
To meet the rigorous demands of mission-critical environments, Advantech participates in the NVIDIA Halos AI Systems Inspection Lab and has partnered with FORT Robotics (FORT) to seamlessly integrate their embedded safety architecture. NVIDIA is the first company in the world to establish an ANSI National Accreditation Board (ANAB)-accredited AI Systems Inspection Lab, integrating functional safety, cybersecurity, AI, and regulations into a unified safety framework. This collaboration ensures that robotic and AMR applications can operate with deterministic performance and SIL-certified dependable control.
NVIDIA Halos is a full-stack comprehensive safety system for physical AI that unifies safety elements from vehicle and robotic architecture to AI models. It comprises HW and SW elements, tools, models, and design principles for combining them to safeguard AI-based, end-to-end AV and robotic stacks.
“Physical AI deployments require more than raw compute – they require integrated safety and control,” said Magic Pao, Associate Vice President at Advantech. “By collaborating with FORT Robotics and leveraging the NVIDIA IGX Thor platform, we deliver a unified foundation that simplifies integration while meeting the highest standards for performance and safety.”
“Safety must be foundational, not an afterthought,” said Amod Damle, Head of Product at FORT. “By embedding functional safety directly into the Edge AI platform, our collaboration with Advantech gives robotics developers the confidence to scale in complex, dynamic environments – unlocking safe, productive operation across a variety of applications.
Unprecedented Edge AI Computing with NVIDIA IGX Thor
The MIC-735 is a production-ready Edge AI inference system delivering up to 2,070 FP4 TFLOPS for deterministic AI in safety-critical environments. Advantech embeds functional safety into its platforms to support safe human-machine collaboration and leverages NVIDIA Halos AI Systems Inspection Lab to inspect the implementation and integration of NVIDIA IGX safety requirements. The system incorporates both onboard and external sensor data to enhance safety concepts and bring physical AI from concept to deployment.
Integrated functional safety architecture
The MIC-735 is built on NVIDIA IGX Thor solid reference design with integrated functional safety. It also supports direct integration with the FORT Nano Safety Controller Pro (NSC Pro). This embedded control and safety receiver enables ecosystem partners who want to apply outside-in robot functional safety to their fleet, ensuring that external safety commands are securely received and executed by the machine’s internal architecture to mitigate risks in real-time.
The MIC-735 also supports a 10-year lifecycle and NVIDIA AI Enterprise-IGX software stack, enabling customers to accelerate applications using NVIDIA NIM microservices, NVIDIA Isaac’s openmodels, libraries, and frameworks for robotics, NVIDIA Metropolis for vision AI, and NVIDIA Holoscan for real-time sensor processing. For functional safety management, the FORT ecosystem provides an API for dynamic configuration changes, allowing developers to integrate safety management directly into their existing enterprise systems.
Sensor synchronisation & rugged connectivity
By integrating NVIDIA Holoscan platform including Holoscan Sensor Bridge technologies for real-time sensing, the MIC-735 enables high-bandwidth, synchronised communication between multiple sensors and AI processors. The system ensures reliable operation from –30 °C to +60 °C and offers high-speed connectivity including 5G and SFP28 4×25 GbE. On the safety side, inputs are securely routed through the FORT Endpoint Controller (EPC), which interfaces with the embedded NSC Pro via specific protocol.
Designed for real-world robotics deployment
Built for deployment at scale in real-world conditions, the MIC-735 combined with FORT technology is ideal for robotic and AMR applications. It is specifically designed for environments navigating complex terrain that require real-time, certified E-Stop capabilities via secure radio or IP networks, utilising devices like the Endpoint Controller to send fleet-wide commands, or the Safe Remote-Control Pro and Wireless E-Stop Pro for local control or command of single AMRs as needed.
By combining NVIDIA AI infrastructure with FORT Robotics’ compact safety footprint, Advantech provides a complete foundation with a long lifecycle that significantly reduces time-to-market and integration complexity for physical AI deployments. The MIC-735 unifies rugged industrial design, NVIDIA IGX Thor performance, and integrated functional safety in a single, deployment-ready system. Together, these capabilities establish a new benchmark for safe, real-time Edge computing in robotics and autonomous systems.
Advantech to showcase at NVIDIA GTC 2026
Advantech invites global partners, developers, and innovators to explore its next-generation edge AI platforms at NVIDIA GTC 2026, taking place 16–19 March in San Jose, California. To learn more about the joint solution from Advantech and FORT Robotics, visit booth #1134 to see safe industrial intelligence in action.
There’s plenty of other editorial on our sister site, Electronic Specifier! Or you can always join in the conversation by visiting our LinkedIn page.
