ASUS PE1100N Edge AI System — NVIDIA Jetson Orin NX/Orin Nano

The ASUS PE1100N is a fanless, ultra-compact edge AI computer built on the NVIDIA Jetson Orin platform, engineered for the embedded and robotics engineers who need to deploy computer-vision and inferencing workloads into vehicles, kiosks, production lines, and outdoor infrastructure — not a data-centre rack. If your project needs real AI throughput running unattended in a cabinet, a vehicle, or a weatherproof enclosure for years at a time, this is a purpose-built industrial edge box, not a repurposed consumer mini PC.

Key Features

• NVIDIA Jetson Orin Module, Configurable to Your Workload — available with Orin Nano 4GB (6-core Arm Cortex-A78AE, 512-core Ampere GPU, 16 Tensor Cores), Orin Nano 8GB (6-core, 1024-core GPU, 32 Tensor Cores), Orin NX 8GB, or Orin NX 16GB (both 8-core, 1024-core GPU, 32 Tensor Cores) — so you only pay for the compute headroom your model actually needs.
• Up to 100 TOPS of AI Inferencing — the top-spec Orin NX 16GB configuration in Super Mode reaches roughly 100 TOPS, with the Nano 4GB/8GB and NX 8GB variants scaling down from there, all inside a 10–25W power envelope and delivering up to 18x better performance-per-watt than a comparable x86 platform.
• Fanless, MIL-STD-810H-Tested Chassis — a 152 x 114 x 72mm extruded-aluminium enclosure weighing 1.4kg, shock- and vibration-tested to MIL-STD-810H Method 514.8/516.8, built to survive vehicle mounting and industrial vibration, not just a lab bench.
• Serious Industrial I/O — 2x Gigabit Ethernet, 3x USB 3.2 Gen1 Type-A, 2x RS-232/422/485 serial (DB9), 1x CAN bus (DB9), and an isolated 4x DI / 4x DO terminal block for sensors, PLCs, and actuators — the connectivity a factory floor or AGV actually needs, not an afterthought.
• Modular Expansion via M.2 — M-key slot for NVMe storage (128GB/256GB/512GB options), E-key for Wi-Fi/Bluetooth, and B-key with dual nano-SIM for 4G/5G cellular — build the exact connectivity loadout your deployment requires.
• Wide-Range Power & Extreme Temperature Tolerance — runs from 12–24V DC (3-pin terminal block, with an optional 65W AC adapter for bench use) across a –20°C to 50°C operating range, suited to vehicle electrical systems, kiosks, and unheated industrial cabinets.
• AWS IoT Greengrass Certified — validated for cloud-connected edge AI deployments where the device needs to sync models and telemetry back to a fleet-management platform.
• TPM 2.0 & Hardware Watchdog — onboard security and automatic recovery from software hangs, both standard requirements for unattended field deployments.

Built for Real Deployments, Not Just Benchmarks

Most Jetson carrier boards on the market are development kits dressed up for production — this is the opposite. The PE1100N ships in a sealed, fanless aluminium enclosure with MIL-STD-810H shock and vibration certification, a wide 12–24V DC input tolerant of the voltage sag and spikes common in vehicle and industrial power systems, and an operating range that covers everything from an unheated warehouse to a hot rooftop enclosure. That matters because the failure mode for edge AI hardware is rarely “not enough TOPS” — it’s a fan that dies after eight months, a board that browns out on a forklift’s electrical system, or a chassis that can’t survive being bolted to a moving vehicle. ASUS designed this one to not be the weak link.

The module lineup lets you right-size compute to the job instead of overpaying for headroom you won’t use. A single-camera people-counting or license-plate-recognition application can run comfortably on the Orin Nano 4GB or 8GB, while multi-camera tracking, real-time object detection at higher resolution, or heavier vision-transformer models justify stepping up to the Orin NX 8GB or 16GB — where AI throughput scales up to roughly 100 TOPS. Whichever module you choose, the software stack is the same NVIDIA JetPack/L4T-based Ubuntu environment used across the entire Jetson family, including support for NVIDIA Isaac ROS and ROS2 — so code, models, and container images developed on a Jetson dev kit or a larger Jetson AGX system carry over with minimal rework.

Being direct about scope: this is an inferencing and computer-vision deployment platform, not a training rig — you develop and fine-tune models elsewhere and deploy the resulting optimized model to the PE1100N for low-latency, offline inference at the point of capture. That’s exactly the workload edge AI in transportation, manufacturing, and robotics actually needs: keeping raw camera and sensor data local instead of streaming it to the cloud, cutting latency to milliseconds, and continuing to operate even when connectivity drops. Typical deployments ASUS targets include traffic and vehicle analysis, people-counting and surveillance, automated guided vehicles (AGVs) and autonomous mobile robots (AMRs), and automated optical inspection (AOI) on production lines — anywhere a camera needs a decision made in real time, locally.

Specifications

• Compute Module: NVIDIA Jetson Orin Nano 4GB / Orin Nano 8GB / Orin NX 8GB / Orin NX 16GB
• CPU: 6-core Arm Cortex-A78AE (Orin Nano) or 8-core Arm Cortex-A78AE (Orin NX)
• GPU: 512-core NVIDIA Ampere with 16 Tensor Cores (Nano 4GB) or 1024-core Ampere with 32 Tensor Cores (Nano 8GB / NX 8GB / NX 16GB)
• Module Memory: 4GB, 8GB, or 16GB LPDDR5 (64-bit or 128-bit, module-dependent)
• AI Performance: Up to 100 TOPS (Orin NX 16GB, Super Mode)
• Storage: M.2 2280 M-key NVMe SSD, 128GB / 256GB / 512GB options
• Display: 1x HDMI (1.4a, 4K30 on Orin Nano SKUs; 2.0b, 4K60 on Orin NX SKUs)
• Networking: 2x Gigabit Ethernet (RJ-45), M.2 E-key 2230 for Wi-Fi/Bluetooth, M.2 B-key 3042/3052 for 4G/5G with dual nano-SIM
• USB: 3x USB 3.2 Gen1 Type-A, 1x USB 2.0 Micro-USB (OS flash/debug), 2x USB 2.0 internal header
• Serial & Fieldbus: 2x RS-232/422/485 (DB9), 1x CAN bus (DB9)
• Digital I/O: 4x isolated DI + 4x isolated DO (terminal block)
• Security: TPM 2.0, hardware watchdog timer
• Power Input: 12–24V DC, 3-pin terminal block; optional 65W AC adapter (100–240VAC in, 19VDC/3.42A out)
• System Power Draw: Approx. 10–25W
• Operating Temperature: –20°C to 50°C
• Storage Temperature: –40°C to 85°C
• Humidity: 10–95% non-condensing
• Shock / Vibration: MIL-STD-810H Method 516.8 (shock) / Method 514.8 (vibration)
• Dimensions: 152 x 114 x 72mm
• Weight: 1.4kg
• Chassis: Fanless extruded-aluminium, wall/DIN-rail mountable
• Operating System: Ubuntu Desktop (NVIDIA JetPack / L4T), with NVIDIA Isaac ROS / ROS2 support
• Certifications: CE, FCC, RoHS; AWS IoT Greengrass Qualified Device