dji-sdk/Payload-SDK

DJI Payload SDK Official Repository

The DJI Payload SDK (PSDK) is an official development kit from DJI for building payloads — add-on devices that attach to DJI drones. It works with DJI's adapter hardware (X-Port, SkyPort, or extension port adapter) so your custom device can exchange information and resources with the drone itself.

Developers use it to write the software logic that drives their payload. The README lists example use cases:

  • Automated Flight Controllers — custom autopilot-style control
  • Payload Controllers — operate cameras, sensors, or other gear
  • Video Image Analysis Platforms — onboard video processing and analysis
  • Mapping Cameras — aerial surveying and reconstruction
  • Megaphones and Searchlights — public-address and illumination tools

The codebase itself is MIT-licensed and targets Linux and RTOS platforms. Full documentation and API references live on DJI's developer site, and support is available via DJI's developer forums, a support ticket system, and Stack Overflow (tag: dji-sdk). The project is actively maintained — the version badge shows V3.17.0 — and contributions are welcome, though DJI notes pull-request review may not be fast, with questions routed to dev@dji.com.

In short: it's the bridge that lets third-party hardware creators make their own devices work smartly with DJI drones.

Related

  • Project
    MIT-SPARK/DAAAM

    DAAAM is a real-time robot mapping system that uses foundation models to build 3D Dynamic Scene Graphs, enabling robots to describe objects and their locations in dynamic environments.

  • Project
    gazebosim/ros_gz

    A set of integration packages that connect the ROS 2 robotics framework with the Gazebo simulator, enabling bidirectional data exchange and robot control.

  • Project
    MIT-SPARK/Khronos

    A spatio-temporal metric-semantic SLAM system that enables robots to map dynamic environments by reasoning about both short-term movements and long-term scene changes.

  • Project
    hustvl/Senna

    Senna is an autonomous driving system that integrates a Large Vision-Language Model with an end-to-end model to enhance planning safety, robustness, and generalization.