Create a KML file with equally spaced points at vertices of a triangular grid.

triangular_grid.py

#!/usr/bin/python3

import pyproj
import simplekml
'''
Generates a KML file with equally spaced points at vertices of a triangular grid. The epsg argument is the code for the UTM 
projection which converts lat/lon to meters. The epsg default value, EPSG:32655 is for Guam (zone 55n).

Example: This command line places points equally spaced at 10 m on the lawn in front of the Canadian Parliament Building:
$ python -c 'from triangular_grid import *; \
generate_equidistant_points(epsg="EPSG:32618", ll_longitude=-75.698873, ll_latitude=45.424078, meters_between_points=10)'
'''
def generate_equidistant_points(epsg='EPSG:32655', meters_between_points=100.0,
    ll_longitude=144.71318, ll_latitude=13.26232, ncols=5, nrows=5,
    kml_file='points.kml'):

    wgs84=pyproj.Proj(init='EPSG:4326') # LatLon with WGS84 datum used by GPS units and Google Earth
    utm=pyproj.Proj(init=epsg) # UTM coords (Cartesian coordinates in meters)
    hdist = meters_between_points # horizontal distance between points
    vdist = meters_between_points * (0.75**0.5) # vertical distance between points
    x0, y0 = pyproj.transform(wgs84, utm, ll_longitude, ll_latitude) # convert lon lat to Cartesian coordinates in meters (UTM)
    x = x0; y = y0; i = 0;  kml = simplekml.Kml()
    for row in range(nrows):
        y = y0 + row * vdist
        for col in range(ncols):
            x = x0 + col * hdist
            if row % 2 != 0: # Odd numbered rows are offset by dist/2
                x = x + hdist / 2
            lon, lat = pyproj.transform(utm, wgs84, x, y) # convert UTM back to lon lat
            i = i + 1
            kml.newpoint(name=str(i), coords=[(lon, lat)])
    kml.save(kml_file)
    return

Thank you very much from Brazil. In Brazil is defaulf use coordinates UTM.