State-Estimation

Practice programs for simple state estimation problems in robotics.

Reference: State Estimation for Robotics (2020) by Timothy D. Barfoot.

Link to YouTube playlist:

https://www.youtube.com/watch?v=jsQ5yIhPXb4&list=PLOxq1EUcxzrKmPDE9FOWAIaEXC9lrWu0h&ab_channel=GlennShimoda

Implement Code

Clone the repository:

git clone https://github.com/TakShimoda/ME8135-State-Estimation.git

Create a virtual environment:

pip install virtualenv

cd ME8135-State-Estimation

virtualenv ME8135

ME8135\Scripts\activate

Install the dependencies:

install -r requirements.txt

Assignments Overview

HW1: Basic Kalman Filter for a linear motion robot. Uses linear motion and observation models and the robot moves in a straight line diagonally.

YouTube Link:

HW2: Extended Kalman Filter, using both the linear measurement model from HW1 and a new measurement model that uses distance and bearing between the robot and a landmark.

YouTube Links for simulations with covariance ellipses:

HW3: Particle Filter, using the same same set up from HW2 (linear measurement model and bearing between robot and a landmark)

Initial particles in blue. Red is the ground truth(robot motion with no noise)

After particles converged.

YouTube Links for simulations with linear and nonlinear measurements:

Part 1: Linear Measurements

Part 2: Nonlinear Measurements