Release Version 6.0.0

Version 6.0.0 refactors the StateType enum in state representation to be more descriptive and widely useful as well as 
the dynamical systems factory for easier creation of these classes. This major change breaks implementations using
StateType enums explicitly (e.g. for parameters) and dynamical systems from the prior version.

See the updated documentation for usage guidelines for the DynamicalSystemsFactory.

This release also includes a few new features, fixes and improvements as well as updated documentation in 
all parts of the control libraries.

CHANGELOG.md

@@ -1,6 +1,7 @@
 # CHANGELOG
 
 Release Versions:
+- [6.0.0](#600)
 - [5.2.0](#520)
 - [5.1.0](#510)
 - [5.0.0](#500)
@@ -11,6 +12,65 @@ Release Versions:
 - [2.0.0](#200)
 - [1.0.0](#100)
 
+## 6.0.0
+
+Version 6.0.0 refactors the StateType enum in state representation to be more descriptive and widely useful as well as 
+the dynamical systems factory for easier creation of these classes. This major change breaks implementations using
+StateType enums explicitly (e.g. for parameters) and dynamical systems from the prior version.
+
+See the updated documentation for usage guidelines for the DynamicalSystemsFactory.
+
+This release also includes a few new features, fixes and improvements as well as updated documentation in 
+all parts of the control libraries.
+
+### Breaking changes
+
+This release contains the following breaking changes:
+- Refactor `StateType` enum and `ParameterType` enum (#277, #278, #280, #284)
+- Relocate the `DYNAMICAL_SYSTEM_TYPE` enum to the general `dynamical_systems` namespace (#288)
+
+**state_representation**
+
+To make the `StateType` enum more descriptive and useful, individual spatial state types have been added
+and the `ParameterType` moved to a separate enum. This change helps with the translation of parameters and
+encoded states in downstream projects by allowing for greater introspection of data types from this enum.
+
+Any implementations that use `StateType` from a prior version to create parameters or compare state types should be
+updated.
+
+**dynamical_systems**
+
+The previous `DYNAMICAL_SYSTEM` enum has been removed from the dynamical system factory class and put directly in the
+general `dynamical_systems` namespace and renamed to `DYNAMICAL_SYSTEM_TYPE` to shorten the creation and to have the
+same structure as the controllers factory. 
+
+Any implementations that use the previous enum have to be updated. See the documentation for more information.
+
+### Features
+
+**state_representation**
+- Add CartesianState attribute setters from std vector and coefficients (#291)
+
+**python**
+- Use pyquaternion in bindings (#283)
+
+**protocol**
+- Add encode/decode options for shared pointer of State (#287)
+- Add CMake config for clproto (#292)
+
+### Fixes and improvements
+
+**state_representation**
+- Mark Parameter getters as const (#289)
+- Throw exception upon parameter validation in controllers (#290)
+
+**python**
+- Fix if else conditions in setup.py to correctly install modules (#285)
+
+**general**
+- Update and extend demos with Python examples and migrate ROS demos away (#293)
+- Update documentation (#294)
+
 ## 5.2.0
 
 Version 5.2.0 contains a few fixes and a new feature for the Impedance controller.

README.md

@@ -12,7 +12,7 @@
         </td>
     </tr>
     <tr>
-        <td width="25%"><a href="https://github.com/epfl-lasa/control-libraries/tree/develop">Development</a></td>
+        <td width="25%"><a href="https://github.com/epfl-lasa/control-libraries/tree/develop">Develop</a></td>
         <td width="75%"><img src="https://github.com/epfl-lasa/control-libraries/actions/workflows/build-test.yml/badge.svg?branch=develop"></td>
     </tr>
 </table>
@@ -21,18 +21,30 @@
 A set of libraries to facilitate the creation of full control loop algorithms,
 including trajectory planning, kinematics, dynamics and control.
 
-Documentation is available at <a href="https://epfl-lasa.github.io/control-libraries/versions/main/">epfl-lasa.github.io/control-libraries</a>
+Documentation is available at <a href="https://epfl-lasa.github.io/control-libraries/versions/">epfl-lasa.github.io/control-libraries</a>.
 
 ## Core libraries
 
-For the implementation, installation and documentation of the core libraries, see the
-<a href="https://github.com/epfl-lasa/control-libraries/tree/main/source">source</a> folder.
+For the implementation, installation and documentation of the core libraries, see the [source](./source) folder.
+
+## Protocol
+
+There is a module that defines the protocol for sending and receiving messages containing control libraries
+data across any network, based on the Google Protocol Buffer. For its implementation, installation and
+documentation, see the [protocol](./protocol) folder.
 
 ## Python bindings
 
-There exist Python bindings for core control library modules. See the <a href="https://github.com/epfl-lasa/control-libraries/tree/main/python">python</a>
-folder for installation instructions and currently supported libraries.
+There exist Python bindings for the control library modules and the protocol module. See the [python](./python)</a>
+folder for installation instructions.
 
 ## Demos
 
-For examples and demos in plain C++, ROS, and ROS2, refer to the <a href="https://github.com/epfl-lasa/control-libraries/tree/main/demos">demos</a> folder.
+For examples and demos in C++ and Python, refer to the [demos](./demos) folder.
+TODO link ros demos repo
+
+## External resources
+
+- C++ remote development in CLion [here](https://github.com/eeberhard/docker-clion-cpp-env)
+- ROS and ROS2 demos using control libraries [here](https://github.com/domire8/control-libraries-ros-demos)
+- ROS and ROS2 control libraries images [here](https://github.com/aica-technology/docker-images)

VERSION

@@ -1 +1 @@
-5.2.0
+6.0.0

demos/control_loop_examples/CMakeLists.txt → demos/CMakeLists.txt

@@ -26,13 +26,10 @@ set(DEMOS_SCRIPTS
 )
 
 set(FIXTURE_INSTALL_PATH /etc/fixtures/)
-install(DIRECTORY ${CMAKE_CURRENT_SOURCE_DIR}/scripts/fixtures/ DESTINATION ${FIXTURE_INSTALL_PATH})
+install(DIRECTORY ${CMAKE_CURRENT_SOURCE_DIR}/fixtures/ DESTINATION ${FIXTURE_INSTALL_PATH})
 
 foreach(SCRIPT ${DEMOS_SCRIPTS})
-  add_executable(${SCRIPT} scripts/${SCRIPT}.cpp)
+  add_executable(${SCRIPT} cpp_scripts/${SCRIPT}.cpp)
   target_link_libraries(${SCRIPT} ${control_libraries_LIBRARIES})
   target_compile_definitions(${SCRIPT} PRIVATE SCRIPT_FIXTURES="${FIXTURE_INSTALL_PATH}")
-  install(TARGETS ${SCRIPT}
-    RUNTIME DESTINATION bin
-  )
 endforeach()

demos/Dockerfile

@@ -0,0 +1,16 @@
+FROM ghcr.io/epfl-lasa/control-libraries/development-dependencies
+
+WORKDIR /tmp
+ARG CONTROL_LIBRARIES_BRANCH=develop
+RUN git clone -b ${CONTROL_LIBRARIES_BRANCH} --depth 1 https://github.com/epfl-lasa/control-libraries.git
+RUN cd control-libraries/source && ./install.sh --auto
+RUN cd control-libraries/protocol && ./install.sh --auto
+RUN pip3 install control-libraries/python
+
+RUN rm -rf /tmp/*
+
+USER developer
+WORKDIR ${HOME}/control_loop_examples
+COPY ./ ./
+
+RUN mkdir build && cd build && cmake .. && sudo make -j all && sudo make install

demos/README.md

@@ -0,0 +1,94 @@
+# `control_loop_examples` demonstration scripts
+
+## Table of contents:
+* [Running a demo script](#running-demonstration-scripts)
+* [Task Space control loop example](#task_space_control_loop)
+* [Robot kinematics control loop example](#robot_kinematics_control_loop)
+
+## Running demonstration scripts
+This package contains a set of demonstration scripts in both C++ and Python that showcase the functionalities introduced
+in the different libraries of `control-libraries`.
+The easiest way to run them is to use the `run-demo.sh` file.
+Without arguments, this script creates a container and opens in interactive mode, allowing you to browse the different
+demo scripts and run the one of your choice:
+
+```console
+./run-demo-script.sh
+# Run a python script
+developer@xxxxxxxxx:~/control_loop_examples$ python3 python_scripts/<script>.py
+# Run a cpp script from the build folder
+developer@xxxxxxxxx:~/control_loop_examples$ ./build/<script>
+```
+
+The available scripts are:
+
+### `task_space_control_loop`
+- **Showcased libraries:** `state_representation`, `dynamical_systems`
+
+This simple demonstration shows how to create a control loop with a `PointAttractor` dynamical system in task space
+(`CartesianState`). It moves a pose towards a random attractor in a `100Hz` control loop.
+The script outputs the current pose and distance to the attractor at each timestep, and the final pose on reaching it:
+
+```console
+...
+frame CartesianPose expressed in world frame
+position: (0.596701, 0.822976, -0.604581)
+orientation: (0.246491, -0.314866, -0.896822, 0.189239) <=> theta: 2.64348, axis: (-0.32489, -0.925375, 0.195264)
+distance to attractor: 0.001004
+-----------
+frame CartesianPose expressed in world frame
+position: (0.596702, 0.822979, -0.604584)
+orientation: (0.246488, -0.314866, -0.896823, 0.189239) <=> theta: 2.64348, axis: (-0.324891, -0.925375, 0.195264)
+distance to attractor: 0.000994
+-----------
+##### TARGET #####
+frame CartesianPose expressed in world frame
+position: (0.59688, 0.823295, -0.604897)
+orientation: (0.246242, -0.314924, -0.896867, 0.189256) <=> theta: 2.64399, axis: (-0.324929, -0.92536, 0.195269)
+##### CURRENT POSE #####
+frame CartesianPose expressed in world frame
+position: (0.596702, 0.822979, -0.604584)
+orientation: (0.246488, -0.314866, -0.896823, 0.189239) <=> theta: 2.64348, axis: (-0.324891, -0.925375, 0.195264)
+```
+
+### `robot_kinematics_control_loop`
+- **Showcased libraries:** `state_representation`, `dynamical_systems`, `robot_model`
+
+This demonstration reuses the previous dynamical system control loop but adds the robot component.
+The desired command, i.e. desired twist of the robot end-effector (eef) is sent to a dummy robot.
+At each timestep, the new joint state and eef state of the robot is computed using the robot kinematics from the robot
+model corresponding to the provided URDF.
+The robot is assumed to perfectly follow the computed desired state matching the eef desired twist.
+Similarly to the previous demonstration, the script outputs the current joint positions, eef pose and distance to the
+attractor at each timestep, and the final joint positions and pose on reaching it:
+
+```
+franka JointPositions
+names: [panda_joint1, panda_joint2, panda_joint3, panda_joint4, panda_joint5, panda_joint6, panda_joint7, ]
+positions: [-1.39453, 0.34338, 1.78372, -1.21918, -0.365912, 1.1687, -2.88819, ]
+panda_link8 CartesianPose expressed in panda_link0 frame
+position: (0.500692, 4.30377e-05, 0.750494)
+orientation: (-9.36521e-05, 8.70069e-05, 1, -0.000152405) <=> theta: 3.14141, axis: (-8.70069e-05, -1, 0.000152405)
+distance to attractor: 0.001249
+-----------
+franka JointPositions
+names: [panda_joint1, panda_joint2, panda_joint3, panda_joint4, panda_joint5, panda_joint6, panda_joint7, ]
+positions: [-1.39627, 0.352394, 1.79511, -1.2227, -0.374677, 1.16757, -2.88157, ]
+panda_link8 CartesianPose expressed in panda_link0 frame
+position: (0.500192, 1.07333e-05, 0.750183)
+orientation: (-1.9014e-05, 4.77102e-05, 1, -7.96025e-05) <=> theta: 3.14155, axis: (-4.77102e-05, -1, 7.96025e-05)
+distance to attractor: 0.000455
+-----------
+##### TARGET #####
+panda_link8 CartesianPose expressed in panda_link0 frame
+position: (0.5, 0, 0.75)
+orientation: (6.12323e-17, 0, 1, 0) <=> theta: 3.14159, axis: (0, 1, 0)
+##### CURRENT STATES #####
+franka JointPositions
+names: [panda_joint1, panda_joint2, panda_joint3, panda_joint4, panda_joint5, panda_joint6, panda_joint7, ]
+positions: [-1.39215, 0.362964, 1.80315, -1.22376, -0.385704, 1.16489, -2.87388, ]
+panda_link8 CartesianPose expressed in panda_link0 frame
+position: (0.500192, 1.07333e-05, 0.750183)
+orientation: (-1.9014e-05, 4.77102e-05, 1, -7.96025e-05) <=> theta: 3.14155, axis: (-4.77102e-05, -1, 7.96025e-05)
+
+```