JETSON & FPGA Heterogeneous System: Camera Link & CoaXPress

Jetson + FPGA

NVIDIA Jetson is a series of embedded computing platforms launched by NVIDIA for edge artificial intelligence (EdgeAI) and robotics development. It combines high-performance GPU, CPU, dedicated AI accelerators with low-power design, specifically built for running AI applications in real-time on the device side (rather than the cloud). It is widely used in robotics, unmanned aerial vehicles (UAVs), smart cameras, industrial automation, medical equipment, autonomous driving cars and other fields. In many scenarios, users need to use GPU to run AI models. As an embedded GPU, Jetson NX has a maximum computing power of 157 TOPS, which can easily run vision models such as YOLO.

JetKU Heterogeneous System

• Core Architecture: Xilinx Kintex UltraScale+ XCKU5P FPGA + NVIDIA Jetson Orin Nano/NX AI SoM dual-processor heterogeneous computing architecture
• Acquisition Channels: 4 independent CoaXPress v2.1 inputs, backward compatible with CXP v2.0/v1.1, maximum 12.5Gbps per channel (CXP-12), total bandwidth 50Gbps; or Camera Link Full interface input
• PoCXP Power Supply: Each CXP interface is built with 17W 24V regulated power supply, overcurrent/short circuit automatic protection, no external power supply required for cameras
• Internal Interconnection: PCIe Gen3 x4 connects FPGA and Jetson, theoretical bandwidth 32Gbps, low-latency large data transmission
• On-board Cache: 2GB DDR4 SDRAM on FPGA side; 4GB/8GB/16GB LPDDR4x on Jetson side (differentiated by model)
• AI Computing Power (Ampere Architecture GPU): Orin Nano 4GB (34TOPS, 7-25W) / Orin Nano 8GB (67TOPS, 7-25W) / Orin NX 8GB (117TOPS, 10-40W) / Orin NX 16GB (157TOPS, 10-40W)
• Edge AI Capability: Supports local running of deep learning models without relying on network connection
• Local Storage: Standard M.2 NVMe 256GB SSD, supporting large-capacity image recording and local deployment of AI models
• Industrial I/O: 26pin isolated industrial interface, including RS485, differential LVDS, optocoupler isolated input/output, TTL encoder interface
• Expansion Interfaces: 40Gbps QSFP+ optical port, Gigabit Ethernet, USB3.2, Type-C, DP display output, Mini PCIe wireless slot
• System Environment: Pre-installed Ubuntu 22.04 LTS, complete NVIDIA JetPack AI software stack, supporting TensorRT accelerated inference
• Mechanical Specifications: Embedded computing platform, dimensions 186.00mm × 173.30mm, 12V DC single power supply for the whole machine

Camera Link & CoaXPress Interfaces

With FMC daughter cards, it can easily adapt to different image interfaces