WiFi SLAM on a Raspberry Pi. This is William and Luke's final project for ECE 5725. Our full project report can be found here.
The work here was based heavly on the paper Efficient, generalized indoor WiFi GraphSLAM
pip3 install adafruit-blinka
pip3 install adafruit-circuitpython-lsm9ds1
pip3 install matplotlib
To run the unit tests run the following command from the main repository directory.
python3 -m unittest testing.test_wifislam
- analysis
- data_association.py: Take in raw measurments and prepare them for the WiFi SLAM code
- dead_reckoning.py: Preform the dead reckoning functions nessary for WiFi SLAM along with full dead reckoning
- tft_plots.py: Display the plots that result from WiFi SLAM and Dead Reckoning on the Pi TFT
- wifislam.py: Does what it says on the tin
- data_collection
- lsm9ds1_data_collection.py: Collect IMU data and write it to a CSV file
- wifi_data_collection.py: Collect WiFi data and write it to a CSV file
- images
- Directory of images that are helpful to the project
- testing
- imu_stats.py: Calculate IMU sample statstics
- lsm9ds1_test.py: Test that the IMU is connected and working
- lsm9ds_time_test.py: Script used to charterize max IMU sampling rate
- sample_imu_stats.txt: IMU stats from our system
- sample_wifi_stats.txt: WiFi stats from our system
- test_wifislam.py: WiFi SLAM unit tests
- control.py: Main control function. Run this to collect data, solve SLAM, and display the plots on the piTFT.