RK3576 Evaluation Board

SKU: DFR1218

The RK3576 Core Board is a high-performance AI computing module featuring Rockchip's AIoT octa-core processor RK3576, suitable for a variety of applications including edge language boxes, image recognition, video surveillance, network video recorders, advertising players, and robot applications. It supports AI edge privatization with a 6TOPS NPU and Mali G52 MC3 GPU, running large model languages like Qwen1.8B, ChatGLM3-6B, and LLaMa2-7B, compatible with frameworks such as tensorflow and pytorch. The board offers millisecond-level response for voice interaction and vision inspection, with 4K video codec and triple-screen display capabilities. Additionally, it provides rich expansion interfaces, including dual high-speed channels and over 12 professional interfaces, enabling connection to devices like industrial cameras and IoT sensor arrays, making it ideal for building intelligent systems from smart retail terminals to automated production lines.

  • Docs
  • Tech Specs
  • 1. Getting Started

    Debugging RK3576 supports methods like serial port, network, and ADB. This guide offers step-by-step instructions for each method, enhancing your troubleshooting skills.

    2. RK3576 interface usage

    The RK3576 interface guide explains GPIO, I2C, and UART functionality, with command line instructions for setup and operation. Explore fan control, RS485, CAN, audio/video interfaces, and network/storage options, enhancing project integration. Ideal for developers optimizing the RK3576 in various applications.

    3. RK3576 AI Example

    Discover the RK3576 AI Example, featuring precompiled RKNN demos on Linux. Utilize dual NPU cores for real-time object detection and instance segmentation. Learn to execute demos with specified model paths and camera devices, supporting tasks like speech recognition and text-to-speech. Perfect for developers exploring AI capabilities on the RK3576 platform.

    Specification

    CPU Block Diagram

    CPU Block Diagram

    Basic Parameters

    Parameter Value
    DC2.5 & Terminal Block Power Input 12-24V DC
    CPU Rockchip RK3576
    Quad Cortex A72 MPCore processor
    Quad Cortex A53 MPCore processor
    NPU 6T TOPS for INT8
    GPU Mali G52 MC3
    DRAM 8GB LPDDR5X
    Memory 64GB EMMC 5.1
    UFS 2.2 (Optional)
    Module size 146x102mm

    Pinout

    On-board Function Diagram

    On-board Function Diagram

    Num. Name Details
    1 DC DC12-24V power input, only one power supply can be connected at a time
    2 ETH Gigabit Ethernet port
    3 CAN CAN FD port
    4 RS485 RS485 port
    5 WiFi WiFi and Bluetooth module
    6 MIPI-CSI MIPI-CSI port
    7 MIPI-DSI MIPI-DSI port
    8 Audio 3.5mm audio signal output
    9 MIC MIC port
    10 SPK Power amplifier output port
    11 RTC RTC battery port
    12 SD microSD card port
    13 DP&HDMI DP and HDMI video output ports
    14 USB2 USB2.0 port
    15 USB3 USB3.0-A and USB3.1 Type-C ports
    16 Recovery Loader button: press and hold to power on and enter Loader mode
    17 UART UART port (UART0 serves as the debug port)
    18 FAN FAN port
    19 LED Indicator lights
    > CORE: Core board power indicator
    > USER: User-defined indicator
    > M.2B: M.2 B port connection indicator
    > M.2M: M.2 M port connection indicator
    20 RST Reset port and power on/off port

    On-board Function Diagram

    Num. Name Details
    1 SIM SIM card port
    2 M.2 B M.2 B port, compatible with 4G and 5G network cards
    3 M.2 M M.2 M port, compatible with M.2 M accelerator cards and M.2 solid-state drives

    Was this article helpful?

    TOP