Banana shape distribution

This repository is a Python implementation of The Banana Distribution is Gaussian: A Localization Study with Exponential Coordinates. Particularly, it addresses all the sections of the paper but section VII.

Installation

To run the project install its requirements as follows.

pip3 install -r requirements.txt

Defining distributions in Exp. Coordinates.

To launch the distribution experiment, run the next command.

python3 distribution.py --D=1 --n_trials=10000 --mov=linear

The parameter D corresponds to a noise coefficient, n_trials is the number of times the trajectory is integrated and mov is the type of displacement, either an arc or linear. Results demonstrating the distribution modelled in cartesian and exponential coordinates is presented.

	Cartesian	Exponential
Linear
Arc

Propagating distribution in Exponential coordinates

To run the propagation experiment, execute the next command. Nota that the arguments are the same ones explained before.

python3 propagation.py --D=1 --n_trials=10000 --mov=arc

	Exponential (Monte Carlo)	Exponential (Propagation)
Linear
Arc