DJI Cloud API Example

A minimal example demonstrating how to fetch OSD data from DJI drones via the DJI Cloud API, publish it over MQTT, and expose a FastAPI dashboard.

---

Table of Contents

1. Overview
2. Features
3. Prerequisites
4. Installation
5. Configuration
6. Running EMQX (MQTT)
7. Usage
8. Sample Output
9. Contributing
10. License
11. Contact

---

Overview

This project consists of two main components:

• mqtt_client.py: Connects to an MQTT broker, subscribes to DJI topics (sys/#, thing/#), parses JSON payloads, handles update_topo replies, and prints OSD telemetry.
• server.py: Starts the MQTT client in the background and launches a FastAPI server to serve a static dashboard and expose configuration/health endpoints.

---

Features

• MQTT client with:

  • Automatic subscription to sys/# and thing/# topics.
  • Callbacks for status (update_topo) and OSD messages.
  • Clear logs prefixed with [MQTT] and [OSD].
  • Retry logic for DNS/connect errors.

• FastAPI server with:

  • /api/config endpoint returning DJI & MQTT settings.
  • /api/health endpoint with server timestamp.
  • Static dashboard served at / from public/dashboard.html.
  • Global error handlers for 404 and 500.

---

Prerequisites

• Python 3.9 or higher
• Docker (for local EMQX broker)
• Your PC and DJI Smart Controller on the same local network

---

Installation

1. Clone the repository:

   git clone https://github.com/OmerMersin/dji-cloud-platform-minimal.git
   cd dji-cloud-platform

2. (Optional) Create and activate a virtual environment:

   python -m venv venv
   source venv/bin/activate  # Linux/macOS
   venv\Scripts\activate  # Windows

3. Install dependencies:

   pip install -r requirements.txt

---

Configuration

Rename .env.example to .env and set the following variables:

DJI_APP_ID=<your_dji_app_id>
DJI_APP_KEY=<your_dji_app_key>
DJI_LICENSE=<your_dji_license>
HOST_ADDR=<your_broker_host_or_ip>
MQTT_USERNAME=<mqtt_username>
MQTT_PASSWORD=<mqtt_password>

Note: HOST_ADDR is used by mqtt_client.py to determine the broker address. The MQTT port defaults to 1883.

---

Running EMQX (MQTT)

To run a local EMQX broker via Docker:

docker run -d --name emqx \
  -p 1883:1883 \
  -p 8083:8083 \
  -p 8084:8084 \
  -p 8883:8883 \
  -p 18083:18083 \
  emqx:5.0.20

1. Open the management console: http://localhost:18083/
2. Default credentials: admin / public
3. (Optional) Create additional users.

---

Usage

1. Ensure your DJI Smart Controller and PC are on the same network.

2. Run the server (which also starts the MQTT client):

   python server.py

3. Access the dashboard at: http://localhost:3000/

4. Fetch configuration: GET http://localhost:3000/api/config

5. Check health: GET http://localhost:3000/api/health

---

Sample Output

[MQTT] Connecting to localhost:1883 ...
[MQTT] Connected to localhost:1883 with result code 0
[MQTT] Subscribed to topics: sys/#, thing/#
[MQTT] Received message on topic: thing/product/<sn>/osd
[OSD] Lat: 0, Lon: 0, Height: 16.16, Head: 10.4, Pitch: -0.4, Roll: 0.2
[OSD] Additional data:
{ '61-0-0': { 'gimbal_pitch': 0, 'gimbal_yaw': 117.8, ... },
  'firmware_version': '03.31.0000',
  ... }

---

Contributing

1. Fork the repository.
2. Create a new branch: git checkout -b feature/your-feature.
3. Make your changes and commit: git commit -m 'Add new feature'.
4. Push to your branch: git push origin feature/your-feature.
5. Open a Pull Request.

---

License

This project is licensed under the MIT License. See LICENSE for details.

---

Contact

• Author: Omer Mersin (omermersin02@gmail.com)
• GitHub: https://github.com/OmerMersin/dji-cloud-platform-minimal