Skip to content

Python implementation of Kalman Filter to estimate position based on Accelerometer readings.

Notifications You must be signed in to change notification settings

nasrAnthony/IMU-position-tracker

Repository files navigation

Basic Kalman Filter implementation in Python.

The goal is to estimate the position over time using X, Y, Z accelerometer data.

Run the json_encoder_v2.py to get data via socket/port. This will genereate a csv before running the IMU_motion_tracker.py.

IMU_data.csv already contains data, so IMU_motion_tracker.py can be ran right away to test.

Results of 4 straight up and down movements will be shown.

Overview of general logic flow: image

About

Python implementation of Kalman Filter to estimate position based on Accelerometer readings.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages