Engineering note
Time Synchronization Before Sensor Fusion
How timestamps, calibration and deterministic data paths shape robust camera, thermal and inertial fusion at the edge.
Read the note →Hybrid AI & control · FE-DO-01
ForgeEdge-Duo
Real-time AI and Linux, working side by side
A heterogeneous MCU–MPU platform combining Cortex-M55 AI processing with dual Cortex-A7 Linux and Cortex-M4 real-time control.
Platform overview
Real-time AI and Linux, working side by side
ForgeEdge-Duo separates time-critical work from higher-level applications. The MCU layer handles local inference, preprocessing and low-latency control, while the MPU layer runs Linux applications, networking, data management and system services.
Process close to the signal
Cortex-M55 at 800 MHz and an integrated 600 GOPS NPU support local AI and low-latency preprocessing.
Keep Linux and control together
Dual Cortex-A7 at 650 MHz and Cortex-M4 at 209 MHz provide application and real-time execution domains.
Integrate machines and networks
Dual Ethernet, dual USB High-Speed, CAN and RS-485 connect acquisition, control and host systems.
Technical overview
System specifications
Selected specifications from the ForgeEdge-Duo datasheet. Refer to the document for the full configuration and integration notes.
| Area | Specification |
|---|---|
| AI / MCU layer | Arm Cortex-M55 at 800 MHz; 600 GOPS NPU |
| Linux / control layer | 2 × Cortex-A7 at 650 MHz; Cortex-M4 at 209 MHz |
| Memory | 512 MB DDR3L system memory; 4.2 MB Cortex-M55 SRAM |
| Storage | eMMC and microSD / SD |
| Network and field I/O | 2 × Ethernet; 1 × CAN; 1 × RS-485 |
| USB | 2 × USB High-Speed |
| Camera interfaces | MIPI CSI-2 and DCMIPP / DCMI |
| Power | 5 V DC via USB / Type-C; 2 W typical, 4 W peak |
| Software | Multi-layer RTOS / bare-metal and embedded Linux |
| Cooling | Passive |
Applications
Built around your use case
- Smart machine control
- Vision and sensor preprocessing
- Industrial data acquisition
- Linux gateways with real-time AI
Integration notes
Configuration matters
Memory, interfaces and software integration are selected for the application. The current datasheet revision lists EMC and ESD objectives as targets. It does not specify a final enclosure rating or guaranteed AI application throughput.
Product documentation
Download datasheet ↓ForgeEdge-Duo datasheet
English · PDF · 8.1 MB · Direct download, no registration
Product options and revision-specific integration notes are included in the document.
Related
Continue exploring.
Engineering note
FPGA, NPU or MPU? Partitioning an Edge AI Architecture
A practical guide to partitioning edge AI workloads across FPGA, NPU, MPU and MCU resources by latency, power and software needs.
Read the note →
Edge compute platforms
ForgeEdge
A scalable edge-computing family combining ForgeEdge-Core processing, an Alvexis carrier, field interfaces and passive cooling in BOX or PCB packages.
Explore system →