Edge-AI FarEdge FE100.
Unovie Edge-AI FE100 operates as a far edge box in numerous sectors such as healthcare, automotive, and industrial fields. It functions as a Far Edge Gateway connected to Ethernet networks for efficient low latency application delivery, maintaining less than 3 seconds of latency by being closer to the data sources.
Furthermore, it accommodates comprehensive local private deployment capabilities that cater to AI algorithms with high computing needs including models like Llama3, Gemma2 and Mistral or YOLOv8 object detection models.
The system is fully controlled through an API from the Cloud Control Plane and smartly integrates with other Edge Gateways for shared Tensor Processing Unit (TPU) resource utilization.
Platform Capabilities
-
Modern and loosely coupled microservices architecture that is hardware and operating system agnostic
-
Advanced integration with multiple OT device protocols including Modbus, BACnet, MQTT, OPC UA, BLE, Zigbee, EtherCAT, PROFINET and many more, with a Device SDK to create further connectors
-
Integration with multiple IT and Cloud systems including AWS, Google Cloud IoT, Microsoft Azure, IBM Watson IoT and more, with an Application SDK to create your own
-
Provision of edge analytics, decision making, control and visualization with standard APIs and the capability to easily integrate more Critical security features to preserve the safe running of the platform
-
Advanced system tooling to fast-track the development, testing, deployment and administration of the platform
Edge Intelligence
Application Capabilities. - Asset Tracking - Intelligent Mobile Routing - Plant Floor Automation - Temperature Monitoring - Service Observability Gateway - Anomaly Detection with Incident Event Collection
EdgeAI FE100 supports up to 32 channels of 1080P H.264/H.265 video decoding and 32 channels of 1080P HD video processing (decoding + AI analysis),making it ideal for various AI applications such as face detection and license plate recognition on video streaming.
Specifications
