Tutorials for SCPToolbox.jl by ACL

To run the tutorial notebooks out-of-the-box:

• (Binder)
• (GESIS Binder)

Rendered notebooks and slides

Part	Notebooks	Slides	Description
0	p0_welcome.ipynb	p0_welcome.slides.html	Introduction
1	p1_clp.ipynb	p1_clp.slides.html	Conic linear programs (CLPs)
2	p2_scp.ipynb	p2_scp.slides.html	Simple sequential convex programming (SCP) example
3	p3_dubin.ipynb	p3_dubin.slides.html	Dubin's car trajectory optimization
4	p4_rocket.ipynb		Rocket-landing guidance (self-guided tutorial)
4*	p4_rocket_solutions.ipynb		Rocket-landing guidance (solutions)

To set up a complete local installation (including SCPToolbox.jl) follow these steps:

• Install Julia v1.7.1, Python 3.10, matplotlib 3.5.1, and ffmpeg.
• Clone the master branch of SCPToolbox.jl from here into ~/tmp.

$ cd ~/tmp
$ git clone https://github.com/UW-ACL/SCPToolbox.jl.git
$ cd SCPToolbox.jl
$ git checkout master

• Clone this tutorial repository into ~/tmp and install TeX-related packages for plotting scripts in SCPToolbox.jl:

$ cd ../
$ git clone https://github.com/UW-ACL/SCPToolbox_tutorial.git
$ cd SCPToolbox_tutorial
$ xargs sudo apt-get install <apt.txt

• Activate the tutorial.jl package environment in Julia Pkg REPL, add SCPToolbox.jl as a dependency and precompile:

$ cd src
julia> ]
(@v1.7) pkg> activate ..
(tutorial) pkg> dev ../../SCPToolbox.jl/
(tutorial) pkg> precompile

• Re-activate the base Julia environment and install/build the IJulia package to obtain JupyterLab:

(tutorial) pkg> activate
(@v1.7) pkg> add IJulia
(@v1.7) pkg> build IJulia

• Launch JupyterLab from Julia REPL. (Conda and JupyterLab installation (self-contained in Julia) will be automatically prompted if they don't exist already.)

julia> jupyterlab(dir=pwd())

To generate slides from a Jupyter notebook:

Slide creation rubric

Slide	Menu option	Command
New	Slide Type	Slide
Persistent	Slide Type	Fragment
Hide input only	Add Tag	hide-input
Hide output only	Add Tag	hide-output
Hide entire cell	Slide Type	Skip

Example command

jupyter nbconvert --output-dir='slides' p1_clp.ipynb --to slides --no-prompt --TagRemovePreprocessor.remove_input_tags={\"hide-input\"} --TagRemovePreprocessor.remove_all_outputs_tags={\"hide-output\"} --post serve