Discrete controllers for kite power systems.
This package is part of Julia Kite Power Tools, which consists of the following packages:
Installation of Julia
If you do not have Julia installed yet, please read Installation.
Installation as package
It is suggested to use a local Julia environment. You can create it with:
mkdir myproject
cd myproject
julia --project=.
(don't forget typing the dot at the end), and then, on the Julia prompt enter:
using Pkg
pkg"add KiteControllers#main"
You can run the tests with:
using Pkg
pkg"test KiteControllers"
To add the examples and install the packages needed by the examples, run:
using KiteControllers
KiteControllers.install_examples()
exit()
Installation using git
In most cases -- if you want to modify, tune and understand kite controllers -- it is better to check out this project from git. You can do this with:
git clone https://github.com/aenarete/KiteControllers.jl.git
cd KiteControllers.jl
git checkout v0.2.15
For the checkout command, use the tag of the latest version. It is suggested to create a system image:
cd bin
./create_sys_image --update
cd ..
On Linux you can install the autopilot
GUI app as normal application with the command bin/install
.
You can now start Julia with ./bin/run_julia
and execute the autopilot or any other of the examples with:
include("examples/autopilot.jl")
Clicking on the "RUN" button starts the simulation. When the simulation is finished, you can click on the "OK" button at the top left. After some time a plot of the flight should appear. You can use the drop-down menu on the left to select other plots or the statistics. You can also load or save log files or projects. A project consists of a kite/ physical system, controller settings and a wind condition.
You get a menu with all the examples by typing:
menu()
FlightPathController # represents the flight path controller
FPCSettings # settings of the flight path controller
FlightPathCalculator # calculate the planned flight path
FlightPathPlanner # execute the planned flight path
FPPSettings # settings of the flight path planner
KiteModel # simplified model for unit testing
SystemStateControl # high level state machine, receives commands from the GUI
# and calls FlightPathPlanner and WinchController
FlightPathController as specified in chapter six of the PhD thesis of Uwe Fechner.
FlightPathController
FPCSettings
on_control_command(fpc, attractor=nothing, psi_dot_set=nothing, radius=nothing, intermediate = true)
on_est_sysstate(fpc, phi, beta, psi, chi, omega, v_a; u_d=nothing, u_d_prime=nothing)
on_timer(fpc)
calc_steering(fpc, parking)
The control commands are usually received from the FlightPathPlanner, the output of the model or the system state estimator must call on_est_systate()
each time step.
Flight path control of kite power systems in a turbulent wind environment
The FPCA is a stateless component that calculates the coordinates of the attractor points and turn points that determine the flight path. In addition, it calculates the desired turn radius of the turns.
The Flight Path Planner executes the planned flight path using a state machine.
While the Flight Path Planner is in charge of calculating the flight path during automated power production, this component handles other system states, like fully manual operation, automated winch control combined with manual steering, parking etc.
If you want to modify this package, create a fork on github and checkout the fork using Git.
Then create a system image of all the packages, used by Kitecontrollers itself and the examples by running
the script bin/create_sysimage --update
. If you are on Windows make sure you have git-bash installed because this
is a bash script. You can then launch Julia with the command bin/run_julia
which is fast. Try to run the
test scripts from the folder test to understand what they are doing.
Now you can modify the source code according to your needs, restart Julia, run the relevant test script again and see the effect.
If you are happy, commit your changes and create a pull request. If you have any questions, please ask them at https://discourse.julialang.org/ .
This project is licensed under the MIT License. Please see the below WAIVER in association with the license.
Technische Universiteit Delft hereby disclaims all copyright interest in the package “KiteController.jl” (controllers for airborne wind energy systems) written by the Author(s).
Prof.dr. H.G.C. (Henri) Werij, Dean of Aerospace Engineering
A Methodology for the Design of Kite-Power Control Systems
If you like this software, please consider donating to https://gofund.me/bba01d43 .
- Research Fechner for the scientific background of this code
- The meta package KiteSimulators which contains all packages from Julia Kite Power Tools.
- the packages KiteModels and WinchModels and AtmosphericModels
- the packages KiteViewers and KiteUtils