Summary is AI-generated, newsdesk-reviewed
  • Aetina's Jetson Thor platforms at GTC 2026 revolutionize edge AI with enhanced robotic autonomy.
  • DeviceEdge AIB-AT78/68 supports humanoid robots with up to 2070 FP4 TFLOPS AI performance.
  • SuperEdge AEX-2UA1 edge server advances on-premises AI with NVIDIA MGX architecture, ensuring data sovereignty.

Aetina Corporation is set to unveil its Jetson Thor-powered edge computing platforms and NVIDIA MGX edge servers at the GTC 2026 conference.

These cutting-edge technologies are designed to integrate high-precision 3D perception with enterprise AI agents, enhancing autonomous robots and supporting real-time industrial decision-making at the edge.

Showcasing Next-Generation Edge AI Solutions

At NVIDIA GTC 2026, Aetina will demonstrate its innovative edge AI solutions. The company's showcase will feature the DeviceEdge AIB-AT78/68 series, a supercomputing platform leveraging NVIDIA Jetson Thor, and modular NVIDIA MGX Edge Servers with NVIDIA NIM microservices.

These solutions emphasize the importance of advanced generative AI agents and high-precision 3D vision technology in progressing towards fully autonomous robots and automation systems.

Adapting to Operational Demands

Modern AI applications in automation require robust real-time perception

Modern AI applications in automation require robust real-time perception and an intelligent decision-making component capable of dynamic task management. Aetina's participation at GTC underlines its position as a critical infrastructure provider for autonomous systems, delivering both essential perceptive and cognitive capabilities to drive forward industrial edge-intelligence.

Technological Innovations for Autonomous Systems

The DeviceEdge AIB-AT78/68 series is central to Aetina's presentation, operating with NVIDIA Jetson T5000 and T4000 modules. Intended for humanoid and advanced robotic systems, this platform has already been implemented across sectors such as manufacturing, logistics, and services, enabling robots to perform with enhanced decision-making and AI skills.

Enabling Scalable Physical AI

Delivering up to 2070 FP4 TFLOPS of AI performance, the AIB-AT78/68 is equipped with extensive I/O including QSFP28 (100 Gbps), EtherCAT, and dual 10 GbE interfaces. This infrastructure facilitates sensor fusion and motion control, aiding the deployment of physical AI solutions. The platform integrates NVIDIA Isaac and Vision AI tools, bridging the gap between simulation and real-world applications.

Humanoid Surround Vision Solution

This system uses the AIB-AT78/68 platform paired with GMSL2 technology and stereo cameras

Aetina, in collaboration with Stereolabs, will introduce the Humanoid Surround Vision Solution at the conference. This system uses the AIB-AT78/68 platform paired with GMSL2 technology and stereo cameras for advanced 3D perception and navigation, suitable for industrial settings with its broad temperature range and power-efficient design.

Aetina will also highlight the SuperEdge AEX-2UA1 edge server, based on NVIDIA MGX architecture. This model, enhanced by NVIDIA RTX PRO 6000 Blackwell Server Edition GPUs, incorporates NVIDIA NIM microservices to run large language models at the edge. It provides a secure means for enterprises to execute real-time data analytics and decision-making on premises, catering to environments like warehouses where latency-sensitive operations are crucial.

Building Infrastructure for AI

Aetina's Joe Lo stated, "Aetina is deeply integrating GMSL2 3D visual perception technology with the NVIDIA Jetson Thor platform, while leveraging the outstanding inference capabilities of NVIDIA MGX modular edge servers to build diversified, deployable infrastructure for the physical AI era." This approach not only enhances next-generation robots with environmental perception but also facilitates the development of intelligent systems capable of transitioning from prototypes to multi-scenario deployments.

GTC 2026 attendees are invited to visit Booth #139 for a firsthand experience of these technologies. The AIB-AT78/68 series is expected to become commercially available by late March 2026.

In case you missed it

Responsible AI Adoption Starts With Governance
Responsible AI Adoption Starts With Governance

The eagerness to adopt AI in physical security is increasing as teams want to implement technology solutions for faster, smarter operations. At the same time, the conversations sur...

Solink's AI Agents Boost Efficiency Of Existing Infrastructure With Automation
Solink's AI Agents Boost Efficiency Of Existing Infrastructure With Automation

Deploying artificial intelligence (AI) tools should be seen as a business initiative rather than a technology initiative, says Martin Soukup, CTO of Solink, a cloud-based video sec...

rf IDEAS Supports Gallagher Badge In Apple Wallet
rf IDEAS Supports Gallagher Badge In Apple Wallet

rf IDEAS, a global manufacturer of RFID credential readers, announces that its WAVE ID® readers support Gallagher Employee Badge in Apple Wallet, expanding the range of credent...