NRU-170PPC: Edge AI without the cloud, but with a user interface

NRU-170PPC: Edge AI without the cloud, but with a user interface

Decentralised data processing plays an important role in the ongoing growth of digitalisation and automation in industry. Edge computing is becoming a key technological consideration, particularly when low latency, minimal downtime and high security standards are required. At the same time, there is a growing demand for high-performance artificial intelligence (AI) directly on the production line – without having to go via the cloud. This is exactly where the new NRU-170PPC panel PC from German distributor Acceed comes in: an industrial AI edge computer with an integrated screen that is designed for continuous operation under demanding conditions and sets new standards in computing power, robustness and system integration.

The NRU-170PPC is a fanless embedded system with an integrated 10.1-inch touch panel (1920 x 1200 pixels) that has been specially developed for industrial AI applications in harsh environments. Depending on the model, the heart of the system is either the Jetson Orin NX or a Jetson Orin Nano SoM from Nvidia. With up to 100 sparse TOPS (INT8) of AI inference performance and video encoding capacity for up to eighteen 1080P streams simultaneously, significant AI performance is achieved. In a key difference to classic IPCs or cloud-based AI solutions, the NRU-170PPC enables data to be evaluated directly on site – for example, in manufacturing, quality control or autonomous driving in industrial plants. This not only reduces latency and bandwidth requirements, but also ensures data sovereignty and optimises IT security in sensitive production environments.

The economic advantage of the NRU-170PPC clearly lies in the reduction of ongoing IT costs. Local data processing saves cloud fees, supports shorter response times for fault detection (including predictive maintenance), ensures less downtime in production and an overall higher degree of automation. The high GPU performance allows multiple processes to run in parallel without the need for additional hardware. Integrated visualisation via the touch panel enables direct human-machine interaction – a real advantage in retrofit projects or in mobile applications.

The housing of the NRU-170PPC is made entirely of aluminium, protected against water and dust in accordance with IP66, and designed for continuous operation at temperatures ranging from -25 to +60 °C. Particularly noteworthy is the fanless design with passive cooling via the large-area cooling fin profile – an advantage over sensitive fan systems, which can quickly become a problem in dusty or vibration-intensive environments.

The NRU-170PPC is also application-oriented in terms of connectivity. The NRU-172S-PPC model has four GbE PoE+ ports with M12 connectors, while the NRU-171V-PPC model has six FAKRA-Z connectors for GMSL2 vehicle cameras. Compatible cameras with pre-installed drivers are available as an option. In addition, several USB 3.2 ports, isolated DIOs and serial interfaces for industrial controllers are available. One M.2 socket is available for installing fast NVMe SSDs with up to 2 TB of storage space, and another for LTE/5G modules with dual micro SIMs.

With the two NRU-170PPC models, Acceed is launching a practical and future-proof solution that intelligently combines the requirements for industrial AI inference systems: robustness, computing power, energy efficiency and ease of use in a compact panel PC. Depending on the application scenario, the NRU-170PPC can achieve savings of several thousand pounds per year and system compared to classic system solutions, while simultaneously increasing performance and improving integration capabilities.

As a specialist in industrial communication and intelligent system integration, Acceed (acceed.com) supports the selection and implementation of suitable embedded and AI solutions. With its expertise and clear focus on industrial applications, Acceed offers not only hardware, but also competent advice and support.

 

We use cookies on our website. By continuing to use our website, you consent to the use of cookies. With your consent, we will also use marketing cookies. For more information about our cookies, please read our Privacy Statement.