AI/Inference at the Edge
AI inference workloads at the edge such as computer vision, weapons detection and persistent surveillance can be very challenging. Our engineers have developed the ideal solution to overcome these challenges with the Argos Ruggedized AI Edge System.
The Argos System’s GPU-accelerated design addresses all the requirements of AI inference workloads at the edge. Unlike traditional servers, Argos is engineered to operate in the field, on limited power, in a broad temperature range, while resisting dust and moisture. It can also be customized to meet your specific needs.
- Geospatial Intelligence (GEOINT)
- Computer Vision
- Edge Inference
- Weapons Detection
- Persistent Surveillance
- Digital Twins
- NVIDIA A100 GPU
- Intel Xeon 5318N 24 Core processor
- Up to 256GB memory
- Up to 40TB RAID 5 storage
- 25GB Fiber SFP+ networking
- Custom MIL-SPEC chassis
- Fully sealed and vibration resistant – can operate in or on any military vehicle
- Operates in a wide temperature range: 0° to 50°C (32° to 122°F) – external heatsink fans automatically energize above 40°C (104°F)
Argos powered by HEAVY.AI leverages machine learning algorithms and big data processing to analyze 11 billion data points, from 10 years of global ship traffic, to identify dark ships and smuggling activities. It identifies patterns and anomalies, flagging suspicious behavior using historical and real-time monitoring.
AI can also analyze other factors, including weather patterns, sea conditions, and port activity, to build a comprehensive understanding of maritime activity and identify potential smuggling hotspots.
Supported Software & AI Frameworks
Argos supports most popular frameworks and can be deployed with containers (e.g., Kubernetes), container orchestration, HCI stack, Silicon Mechanics AI Stack or Silicon Mechanics’ Scientific Computing Stack.
Speak to an expert to learn more
"*" indicates required fields
Enabling In-field Artificial Intelligence with a Highly Portable Ruggedized Edge Device
A U.S. government defense contractor needed help supporting complex artificial intelligence workloads in a physical device meant to be deployed in a variety of field settings, including large, public events.
This edge system needed to run artificial intelligence workloads such as computer vision and object detection. The complex design also had to be safe from the impact of environmental hazards (including, temperature, dust, vibration, moisture, etc.) and be able to operate in real world field conditions (such as limited size, ability to operate in broad temperature range, limited power, etc.).
Our team was able to create a custom-engineered ruggedized edge system that protects against environmental conditions and allows for in-field deployment results comparable to current vehicle-borne systems.
The in-field edge device proved so useful to the client that they began integrating the component into other DoD project bids.
At the same time, we created a version of the device, affectionately called Argos after the multi-eyed, “all-seeing” Greek legend and the original use case, suited for broader artificial intelligence deployments in other government agencies and service branches.
Made in USA
We build all systems in our US-based manufacturing facilities. We inspect and test each system to ensure it is 100% operational and optimized to support rapid deployment.
FCC Marking (U.S.), CE Marking (EU),
China Compulsory Certificate Marking
UL61010 and UL62368