2.2.7 - Power Modules

The new RQX-59 AI controller is part of the ROScube-X family and is based on Nvidia’s Jetson AGX Orin module. The SoM (System-on-Module) combines the energy-efficient Cortex-A78AE (64-bit) ARM processor with a GPU featuring up to 2048 CUDA processing units and 64 Tensor cores. This enables AI models for object recognition, depth estimation or semantic segmentation to be processed directly within the vehicle or robot. A power budget starting at around 40 W is likely to be attractive for battery-powered mobile applications, without having to compromise on sophisticated models.

With the NRU-160-FT, German distributor Acceed aims to bring AI computing power to the very places where data is collected and decisions are made. The system can be installed in dust- or waterproof enclosures and uses their walls for heat dissipation, for example in construction machinery, in car parks and on roads, or in autonomous logistics vehicles. With Nvidia’s Jetson Orin AI module, the embedded computer achieves up to 100 INT8-TOPS and can simultaneously process up to 18 Full HD video streams, whilst consuming around 25 Watt of power.

Industrial edge computing is increasingly evolving from a data collector into an autonomous decision-maker. Requirements are changing in OT (Operational Technology) networks. The focus is on deterministic communication, a minimal footprint in the control cabinet, AI performance and energy efficiency. This is where Acceed’s new Nuvo-11531 sets new standards. The book-sized computer combines an exceptionally small, fanless housing with the latest Intel Ultra 200S processors, positioning itself between traditional industrial PLCs and GPU vision computers, but with remarkable I/O density.

It is a well-known fact that industrial applications require computing power at the edge of the network, where data is generated. Real-time AI systems for image processing, quality control, or autonomous machines need to evaluate large amounts of data directly on site. Bypassing cloud architectures saves time and bandwidth and reduces costs. However, edge applications have special requirements: high performance in limited space, constant availability despite dust and vibrations, and high tolerance to ambient temperatures. The new GT-92GC, a 19-inch rack system that combines GPU acceleration, multi-camera connectivity and a fanless chassis design, is tailored precisely to these requirements.

As AI models become increasingly efficient in industry, the demands on hardware in the field are also growing: GPU performance must be mobile, robust and maintenance-free in vehicles, production lines, energy plants and security systems. With the new GT-92RL-H controller, German distributor Acceed now has another edge platform in its portfolio that is designed precisely for this purpose: powerful, fanless, with high I/O density and, thanks to its robust system design (in accordance with EN 50155 / EN 45545), largely independent of infrastructure conditions.