Research algorithm for the orientation of the TEIDESAT-1 satellite
The IMU Adafruit LSM9DS1 9-DOF is used. It containing multiple sensors for taking different measurements:
- Acceleration (m/s^2): 3 accelerometers (XYZ axis)
- Rotation rate (rad/s): 3 gyroscopes (XYZ axis)
- Magnetic flux (G): 3 magnetometers (XYZ axis)
- Temperature (°C): temperature sensor
datasheet: https://www.st.com/resource/en/datasheet/lsm9ds1.pdf
The U-BLOX NEO 6M GNSS module is use for testing purposes and is not the final
Guide to connect NEO 6M GPS to arduino
https://randomnerdtutorials.com/guide-to-neo-6m-gps-module-with-arduino
Due to the high cost of sun sensors, light sensors will be used for the prototype. The VEML7700 model is the one used in this case.
Using an Espressif ESP32 WROOM 32 microcontroller as the board computer. The I2C protocol is used for the comunication between the microcontroller and the IMU,
Pin Conections (SPI protocol):
| Arduino | IMU |
|---|---|
| 3V | VIN |
| G | GND |
| D1 | SCL |
| D2 | SDA |
Used the PlatformIO plugin for VSCode to create the project and upload the attitude calculation program into the arduino board (configuration of the ESP-12E already set up in the project). Libraries requerided are already integrated in the project in the .pio/libdeps/psp12e directory (They can also be found in the PlatformIO Registry):
Main dependencies (look platformio.ini file for more information):
- adafruit/Adafruit LSM9DS1 Library@^2.2.1
- mikalhart/TinyGPSPlus@^1.1.0
- plerup/EspSoftwareSerial@^8.2.0
All the implementations are located in the lib/algorithms directory.
├── img # Documentation about the project
├── lib # INS (Inertial Navigation System) orientation source code
| ├── hardware # Hardware abstraction to make it easier to connect with the algortihms (all sensors are attached and setup in INS.hpp)
| ├── IMUlgortims # IMU orientation algoritms
| ├── sunSensorAlgorithms # Sun Sensor orientation algoritms
| └── utils # Utilities common to the IMU algorithms
├── src # Multiple main programs, one for each algorithm. Hardware sensor data is read and used to feed the algorithm
└── testTo upload the one of the implementations into the arduino board, change the name of the sketches in the src/ directory from main-IMPLEMENTATION-NAME.txt to main.cpp. Every other implementation should end in .txt as only one main file should exist.
UPSat ADCS software modified to be used with arduino instead of Raspberry
File where the algorithm is implemented: main-upsat.txt
ADCS software folder in their repository https://gitlab.com/librespacefoundation/upsat/upsat-adcs-software/-/tree/master/sensor-fusion-test?ref_type=heads
Algorithm based on ACUBESAT´s implementation. This implementation relies mostly in the gyroscopes to calculate the relative movement. https://gitlab.com/acubesat/adcs/adcs-software/business-logic-software
Library implementating different algorithms for the calculation of the orientation with an easy to setup interface. Using the Madgwick algorithm as it is better than the Mahony algorithm.
File where the algorithm is implemented: main-reefwing.txt
Reefwing Library repository https://github.com/Reefwing-Software/Reefwing-AHRS
Adafruit`s fork library for the implementation of the main orientation algoritms (madgick and mahony).
File where the algorithm is implemented: main-adafruit-fork.txt
Adafruit references a fork of their library in one of their tutorial pages:
https://learn.adafruit.com/ahrs-for-adafruits-9-dof-10-dof-breakout/sensor-fusion-algorithms
Base repository https://github.com/adafruit/Adafruit_AHRS/tree/master
Fork repository
https://github.com/PaulStoffregen/MadgwickAHRS (madgwick algorithm)
https://github.com/PaulStoffregen/MahonyAHRS (mahony algorithm)
Implementation based but not following exactly all the steps on the tutorial. Test the output of every sensor and simple sensor fusion. This implementation is not reliable for use in any kind of vehichle but just an aproximation on how the sensors works.
File where the algorithm is implemented: main-trigonometry.txt
tutorial serie for sensor fusion using trigonometry:
https://www.youtube.com/watch?v=2AO_Gmh5K3Q&list=PLGs0VKk2DiYwEo-k0mjIkWXlkrJWAU4L9&index=1&ab_channel=PaulMcWhorter