Skip to content

Commit

Permalink
Merged
Browse files Browse the repository at this point in the history
  • Loading branch information
@DominicDirkx
DominicDirkx committed Jan 17, 2024
2 parents 797f964 + 582f354 commit 8227687
Show file tree
Hide file tree
Showing 3 changed files with 262 additions and 161 deletions.
210 changes: 156 additions & 54 deletions mission_design/earth_mars_transfer_window.ipynb
View file

Large diffs are not rendered by default.

209 changes: 105 additions & 104 deletions mission_design/earth_mars_transfer_window.py
View file
@@ -1,68 +1,92 @@
# Earth Mars transfer window design with Porkchop Plots
'''
# Earth-Mars transfer window design using Porkchop Plots
"""
Copyright (c) 2010-2023, Delft University of Technology
All rigths reserved
This file is part of the Tudat. Redistribution and use in source and
binary forms, with or without modification, are permitted exclusively
under the terms of the Modified BSD license. You should have received
a copy of the license with this file. If not, please or visit:
http://tudat.tudelft.nl/LICENSE.
'''

## IMPORTANT
"""
This example requires the `tudatpy.trajectory_design.porkchop` module.
Please ensure that your version of tudatpy does include this module.
"""

## Context

## Summary
"""
This example shows how the tudatpy porkchop module can be used to choose
an optimal launch and arrival date for an Earth-Mars transfer. By default,
the porkchop module uses a Lambert arc to compute the ΔV required to
depart from the departure body (Earth in this case) and be captured by the
target body (in this case Mars).
This example demonstrates the usage of the tudatpy `porkchop` module to determine an optimal launch window (departure and arrival date) for an Earth-Mars transfer mission.
Users can provide a custom function to calculate the ΔV required for any
given transfer. This can be done by supplying a `callable` (a function)
to the `porkchop` function via the argument
By default, the porkchop module uses a Lambert arc to compute the $\Delta V$ required to depart from the departure body (Earth in this case) and be captured by the target body (in this case Mars).
Users can provide a custom function to calculate the $\Delta V$ required for any given transfer. This can be done by supplying a `callable` (a function) to the `porkchop` function via the argument
function_to_calculate_delta_v
This opens the possibility to calculate the ΔV required for any transfer
accounting for course correction manoeuvres along a planned trajectory.
This opens the possibility to calculate the $\Delta V$ required for any transfer; potential applications include: low-thrust transfers, perturbed transfers with course correction manoeuvres, transfers making use of Gravity Assists, and more.
"""

## How this example is organized
## Import statements
"""
This example consists of 3 sections:
1. The imports, where we import the required modules
2. Data management, where we define the file where the porkchop data will be saved
3. The porkchop itself, which will only be recalculated if the user requests it
These sections are marked with Matlab-style #%% comments. If you use
the VSCode Jupyter extension this will allow you to run each section
in turn and fold the sections to help you navigate the contents of
the file.
The required import statements are made here, starting with standard imports (`os`, `pickle` from the Python Standard Library), followed by tudatpy imports.
"""

#--------------------------------------------------------------------
#%% IMPORT STATEMENTS
#--------------------------------------------------------------------

# General imports
import os
import pickle

# Tudat imports
from tudatpy import constants
from tudatpy.interface import spice_interface
from tudatpy.astro.time_conversion import DateTime
from tudatpy.numerical_simulation import environment_setup
from tudatpy.trajectory_design.porkchop import porkchop, plot_porkchop

#--------------------------------------------------------------------
#%% DATA MANAGEMENT
#--------------------------------------------------------------------
## Environment setup
"""
We proceed to set up the simulation environment, by loading the standard Spice kernels, defining the origin of the global frame and creating all necessary bodies.
"""

# Load spice kernels
spice_interface.load_standard_kernels( )

# Define global frame orientation
global_frame_orientation = 'ECLIPJ2000'

# Create bodies
bodies_to_create = ['Sun', 'Venus', 'Earth', 'Moon', 'Mars', 'Jupiter', 'Saturn']
global_frame_origin = 'Sun'
body_settings = environment_setup.get_default_body_settings(
bodies_to_create, global_frame_origin, global_frame_orientation)

# Create environment model
bodies = environment_setup.create_system_of_bodies(body_settings)

## Porkchop Plots
"""
The departure and target bodies and the time window for the transfer are then defined using tudatpy `astro.time_conversion.DateTime` objects.
"""

departure_body = 'Earth'
target_body = 'Mars'

earliest_departure_time = DateTime(2005, 4, 30)
latest_departure_time = DateTime(2005, 10, 7)

earliest_arrival_time = DateTime(2005, 11, 16)
latest_arrival_time = DateTime(2006, 12, 21)

# To ensure the porkchop plot is rendered with good resolution, the time resolution of the plot is defined as 0.5% of the smallest time window (either the arrival or the departure window):

# Set time resolution IN DAYS as 0.5% of the smallest window (be it departure, or arrival)
# This will result in fairly good time resolution, at a runtime of approximately 10 seconds
# Tune the time resolution to obtain results to your liking!
time_window_percentage = 0.5
time_resolution = time_resolution = min(
latest_departure_time.epoch() - earliest_departure_time.epoch(),
latest_arrival_time.epoch() - earliest_arrival_time.epoch()
) / constants.JULIAN_DAY * time_window_percentage / 100

# Generating a high-resolution plot may be time-consuming: reusing saved data might be desirable; we proceed to ask the user whether to reuse saved data or generate the plot from scratch.

# File
data_file = 'porkchop.pkl'
Expand All @@ -73,55 +97,14 @@
).strip().lower() == 'y'
print()

#--------------------------------------------------------------------
#%% PORKCHOP
#--------------------------------------------------------------------

# Define departure and target bodies
departure_body = 'Earth'
target_body = 'Mars'
# Lastly, we call the `porkchop` function, which will calculate the $\Delta V$ required at each departure-arrival coordinate and display the plot, giving us
#
# - The optimal departure-arrival date combination
# - The constant time-of-flight isochrones
# - And more

if not os.path.isfile(data_file) or RECALCULATE_delta_v:

# Tudat imports
from tudatpy.interface import spice_interface
from tudatpy.astro.time_conversion import DateTime
from tudatpy.numerical_simulation import environment_setup

#--------------------------------------------------------------------
#%% PORKCHOP PARAMETERS
#--------------------------------------------------------------------

# Load spice kernels
spice_interface.load_standard_kernels( )

# Define global frame orientation
global_frame_orientation = 'ECLIPJ2000'

# Create bodies
bodies_to_create = ['Sun', 'Venus', 'Earth', 'Moon', 'Mars', 'Jupiter', 'Saturn']
global_frame_origin = 'Sun'
body_settings = environment_setup.get_default_body_settings(
bodies_to_create, global_frame_origin, global_frame_orientation)

# Create environment model
bodies = environment_setup.create_system_of_bodies(body_settings)

earliest_departure_time = DateTime(2005, 4, 30)
latest_departure_time = DateTime(2005, 10, 7)

earliest_arrival_time = DateTime(2005, 11, 16)
latest_arrival_time = DateTime(2006, 12, 21)

# Set time resolution IN DAYS as 0.5% of the smallest window (be it departure, or arrival)
# This will result in fairly good time resolution, at a runtime of approximately 10 seconds
# Tune the time resolution to obtain results to your liking!
time_window_percentage = 0.5
time_resolution = time_resolution = min(
latest_departure_time.epoch() - earliest_departure_time.epoch(),
latest_arrival_time.epoch() - earliest_arrival_time.epoch()
) / constants.JULIAN_DAY * time_window_percentage / 100

# Regenerate plot
[departure_epochs, arrival_epochs, ΔV] = porkchop(
bodies,
departure_body,
Expand All @@ -132,32 +115,50 @@
latest_arrival_time,
time_resolution
)
# Save data
pickle.dump(
[departure_epochs, arrival_epochs, ΔV],
open(data_file, 'wb')
)

else:

# Read saved data
[departure_epochs, arrival_epochs, ΔV] = pickle.load(
open(data_file, 'rb')
)
# Plot saved data
plot_porkchop(
departure_body = departure_body,
target_body = target_body,
departure_epochs = departure_epochs,
arrival_epochs = arrival_epochs,
delta_v = ΔV,
threshold = 15
)

### Variations
"""
The Tudat `porkchop` module allows us to
# Cases
cases = [
{'C3': False, 'total': False, 'threshold': 15 , 'filename': 'figures/ΔV.png'},
{'C3': False, 'total': True, 'threshold': 15 , 'filename': 'figures/Δ_tot.png'},
{'C3': True, 'total': False, 'threshold': 42 , 'filename': 'figures/C3.png'},
{'C3': True, 'total': True, 'threshold': 100, 'filename': 'figures/C3_tot.png'}
]

for case in cases:
plot_porkchop(
departure_body = departure_body,
target_body = target_body,
departure_epochs = departure_epochs,
arrival_epochs = arrival_epochs,
delta_v = ΔV,
save = True,
**case
)
- Save the $\Delta V$ map returned by `porkchop` and plot it again without recalculating with the `plot_porkchop` function
- Plot $\Delta V$ (default) or C3 (specific energy), as well as choose whether to plot departure and arrival $\Delta V$ together as the total $\Delta V$ required for the transfer (default), or separately (in those cases in which the manoeuvre is performed in two burns, one at departure and one at arrival to the target planet).
Let's make use of `plot_porkchop` to see all four combinations!
"""

cases = [
{'C3': False, 'total': False, 'threshold': 15 , 'filename': 'figures/ΔV.png'},
{'C3': False, 'total': True, 'threshold': 15 , 'filename': 'figures/Δ_tot.png'},
{'C3': True, 'total': False, 'threshold': 42 , 'filename': 'figures/C3.png'},
{'C3': True, 'total': True, 'threshold': 100, 'filename': 'figures/C3_tot.png'}
]

for case in cases:
plot_porkchop(
departure_body = departure_body,
target_body = target_body,
departure_epochs = departure_epochs,
arrival_epochs = arrival_epochs,
delta_v = ΔV,
save = False,
**case
)
4 changes: 1 addition & 3 deletions propagation/mga_trajectories.ipynb
View file
Expand Up @@ -1129,9 +1129,7 @@
"execution_count": 24,
"id": "4429114f-ea11-45ca-8c16-8568a1e28d46",
"metadata": {
"tags": [
"nbsphinx-thumbnail"
]
"tags": []
},
"outputs": [
{
Expand Down

0 comments on commit 8227687

Please sign in to comment.