Clone this repo
git clone git@github.com:sebastiano123-c/DroneIno.git
or download the .zip file.
This code is intended to give a starting point for your telemetry system. Based on the esp32-cam, this code allow you to:
- see telemetry data of the drone;
- send parameters to the drone;
- stream the camera video;
- store flight data on a SD card.
For example, I use this system on my DroneIno project.
The principle of working is: the drone board is connected via UART (RX/TX pins) to the esp32-cam; Using esp32-cam wifi, it is generated a wifi access point (AP). The connections are shown in the figure below.
transmitter esp32-cam |----------| |----------| | TX |-----------| RX0 | | RX |-----------| TX0 | => wifi => web app | GND |-----------|GND0 | |----------| |----------|
In the wifi-tx example, I use an esp32 board as transmitter. If you don't use a esp32 board, make the proper changes on the code and on the connections.
Connect the RX1 (GPIO 26) of the esp32 to the TX0 (GPIO 1) of esp32-cam, and TX1 (GPIO 25) of the esp32 to the RX0 (GPIO 3) of esp32-cam.
Upload the wifi-tx sketch on your transmitter board. Then, upload the main sketch on your esp32-cam.
After connecting to the network using the password esp32cam, dial in your browser's search bar 192.168.4.1.
That's it.
In the main sketch, I use consider the following variables for telemetry:
- roll
- pitch
- flight mode
- altitude
- battery
- four channels of the controller (e.g. roll, pitch, throttle, yaw)
- GPS latitude, longitude and UTC time. The browser update rate is about 100ms.
PID variables are easily set in the browser in the toggling menu. When you change a certain PID parameter, the latter is sent via WiFi to the esp32-cam, and then back to the transmitter board via UART.
Press the button start video in the toggling menu and the live streaming will appear showing also the status of the telemetry.
You can set the video parameters in the toggling menu.
To better appreciate this telemetry system, I recommend to use also a SD card. SD card must be formatted in FAT32.
Take the SD card, and create a directory and a file called /src/config.txt in which you write only the initial PID parameters.
When the drone starts, the parameters are uploaded to browser and sent back to the drone.
With SD card your flightData will be save in a file in the folder /data/flight_i.csv, which is automatically created.
The directory structure of the SD is
\src
config.txt
\data
\videos
\images
\flightData
flight_1.csv
flight_2.csv
:
flight_n.csv
I also put in the test folder the HTML source. If you modify it, please remember to full compress all css, HTML and js parts, otherwise the memory of the esp32-cam will overflow.
The index file must then be compressed using this gzip procedure, and then put it in the camera index.
To better visualize your flight data, telemetryAnalysis provide a versatile graphic tool for the SD recordings. In telemetryAnalysis folder I put a simple Python script and an executable file working in Windows. The analysis includes:
- roll and pitch angle sequence;
- RC-controller inputs;
- altitude recordings;
- battery recordings. For the python script, I also menage a GPS map.
There few things that must be done in the future:
- save video and photo;
Sebastiano Cocchi