pbdlib-cpp
PbDlib is a collection of C++ source codes for robot programming by demonstration (learning from demonstration). It includes various functionalities at the crossroad of statistical learning, dynamical systems, optimal control and Riemannian geometry.
PbDlib can be used in applications requiring task adaptation, human-robot skill transfer, safe controllers based on minimal intervention principle, as well as for probabilistic motion analysis and synthesis in multiple coordinate systems.
Other versions of the library in Matlab (compatible with GNU Octave) and Python are also available at http://www.idiap.ch/software/pbdlib/ (currently, the Matlab version has the most functionalities).
Usage
This project builds a number of executables in the build folder, with the corresponding C++ source codes in the src folder listed in the table below. The corresponding publications are also listed below.
List of examples
| Filename | ref. | Description |
|---|---|---|
| demo_ergodicControl_2D01.cpp | 5(#ref-5) | 2D ergodic control with spectral multiscale coverage (SMC) algorithm |
| demo_ergodicControl_nD01.cpp | 5(#ref-5) | nD ergodic control with spectral multiscale coverage (SMC) algorithm |
| demo_GMR01.cpp | 5(#ref-5) | Gaussian mixture model (GMM) and time-based Gaussian mixture regression (GMR) used for reproduction |
| demo_GPR01.cpp | 2(#ref-2) | Gaussian process regression (GPR) |
| demo_HSMM_batchLQR01.cpp | 2(#ref-2) | Use of HSMM (with lognormal duration model) and batch LQR (with position only) for motion synthesis |
| demo_infHorLQR01.cpp | 2(#ref-2) | Discrete infinite horizon linear quadratic regulation |
| demo_LWR_batch01.cpp | 2(#ref-2) | Locally weighted regression (LWR) with radial basis functions (RBF), using batch computation |
| demo_LWR_iterative01.cpp | 2(#ref-2) | Locally weighted regression (LWR) with radial basis functions (RBF), using iterative computation |
| demo_MPC_batch01.cpp | 2(#ref-2), 7(#ref-7) | Model predictive control (MPC) with batch linear quadratic tracking (LQT) formulation |
| demo_MPC_iterative01.cpp | 2(#ref-2), 7(#ref-7) | Model predictive control (MPC) with iterative linear quadratic tracking (LQT) formulation |
| demo_MPC_semitied01.cpp | 2(#ref-2), 7(#ref-7) | MPC with semi-tied covariances |
| demo_MPC_velocity01.cpp | 2(#ref-2), 7(#ref-7) | MPC with an objective including velocity tracking |
| demo_online_GMM01.cpp | 2(#ref-2), 6(#ref-6) | Online GMM learning and LQR-based trajectory generation |
| demo_online_HSMM01.cpp | 2(#ref-2), 6(#ref-6) | Online HSMM learning and sampling with LQR-based trajectory generation |
| demo_proMP01.cpp | 5(#ref-5) | Conditioning on trajectory distributions with ProMP |
| demo_Riemannian_pose_batchLQR01.cpp | 3(#ref-3) | Linear quadratic regulation of pose by relying on Riemannian manifold and batch LQR (MPC without constraints) |
| demo_Riemannian_pose_GMM01.cpp | 3(#ref-3) | GMM for 3D position and unit quaternion data by relying on Riemannian manifold |
| demo_Riemannian_pose_infHorLQR01.cpp | 3(#ref-3) | Linear quadratic regulation of pose by relying on Riemannian manifold and infinite-horizon LQR |
| demo_Riemannian_pose_TPGMM01.cpp | 3(#ref-3) | TP-GMM of pose (R3 x S3) with two frames |
| demo_Riemannian_S2_GMM01.cpp | 3(#ref-3) | GMM for data on a sphere by relying on Riemannian manifold |
| demo_Riemannian_S2_infHorLQR01.cpp | 3(#ref-3) | Linear quadratic regulation on a sphere by relying on Riemannian manifold and infinite-horizon LQR |
| demo_Riemannian_S2_product01.cpp | 3(#ref-3) | Gaussian product on sphere |
| demo_Riemannian_S2_TPGMM01.cpp | 3(#ref-3) | TP-GMM for data on a sphere by relying on Riemannian manifold (with single frame) |
| demo_Riemannian_S3_infHorLQR01.cpp | 3(#ref-3) | Linear quadratic regulation on hypersphere (orientation as unit quaternions) by relying on Riemannian manifold and infinite-horizon LQR |
| demo_Riemannian_S3_TPGMM01.cpp | 3(#ref-3) | TP-GMM on S3 (unit quaternion) with two frames |
| demo_Riemannian_SPD_GMR01.cpp | 4(#ref-4) | GMR with time as input and covariance data as output by relying on Riemannian manifold |
| demo_Riemannian_SPD_interp02.cpp | 4(#ref-4) | Covariance interpolation on Riemannian manifold from a GMM with augmented covariances |
| demo_TPGMMProduct01.cpp | 1(#ref-1) | Product of Gaussians for a Task-Parametrized GMM |
| demo_TPGMR01.cpp | 1(#ref-1) | Task-Parametrized GMM with GMR (time as input), the model is able to adapt to continuously changing task parameters. |
| demo_TPMPC01.cpp | 1(#ref-1) | Linear quadratic control (unconstrained linear MPC) acting in multiple frames, which is equivalent to the fusion of Gaussian controllers |
References
If you find PbDlib useful for your research, please cite the related publications!
[1] Calinon, S. (2016). A Tutorial on Task-Parameterized Movement Learning and Retrieval. Intelligent Service Robotics (Springer), 9:1, 1-29.
pdf(http://calinon.ch/papers/Calinon-JIST2015.pdf)
bib(http://calinon.ch/papers/Calinon-JIST2015.bib)
(Ref. for GMM, TP-GMM, MFA, MPPCA, GPR, trajGMM)
[2] Calinon, S. and Lee, D. (2019). Learning Control. Vadakkepat, P. and Goswami, A. (eds.). Humanoid Robotics: a Reference, pp. 1261-1312. Springer.
pdf(http://calinon.ch/papers/Calinon-Lee-learningControl.pdf)
bib(http://calinon.ch/papers/Calinon-Lee-learningControl.bib)
(Ref. for MPC, LQR, HMM, HSMM)
[3] Calinon, S. and Jaquier, N. (2019). Gaussians on Riemannian Manifolds for Robot Learning and Adaptive Control. arXiv:1909.05946.
pdf(http://calinon.ch/papers/Calinon-arXiv2019.pdf)
bib(http://calinon.ch/papers/Calinon-arXiv2019.bib)
(Ref. for Riemannian manifolds)
[4] Jaquier, N. and Calinon, S. (2017). Gaussian Mixture Regression on Symmetric Positive Definite Matrices Manifolds: Application to Wrist Motion Estimation with sEMG. In Proc. of the IEEE/RSJ Intl Conf. on Intelligent Robots and Systems (IROS), pp. 59-64.
pdf(http://calinon.ch/papers/Jaquier-IROS2017.pdf)
bib(http://calinon.ch/papers/Jaquier-IROS2017.bib)
(Ref. for S^+ Riemannian manifolds)
[5] Calinon, S. (2019). Mixture Models for the Analysis, Edition, and Synthesis of Continuous Time Series. Bouguila, N. and Fan, W. (eds). Mixture Models and Applications, pp. 39-57. Springer.
pdf(http://calinon.ch/papers/Calinon_MMchapter2019.pdf)
bib(http://calinon.ch/papers/Calinon_MMchapter2019.bib)
(Ref. for ergodic control, Bezier curves, LWR, GMR, probabilistic movement primitives)
[6] Bruno, D., Calinon, S. and Caldwell, D.G. (2017). Learning Autonomous Behaviours for the Body of a Flexible Surgical Robot. Autonomous Robots, 41:2, 333-347.
pdf(http://calinon.ch/papers/Bruno-AURO2017.pdf)
bib(http://calinon.ch/papers/Bruno-AURO2017.bib)
(Ref. for DP-means)
[7] Berio, D., Calinon, S. and Fol Leymarie, F. (2017). Generating Calligraphic Trajectories with Model Predictive Control. In Proc. of the 43rd Conf. on Graphics Interface, pp. 132-139.
pdf(http://calinon.ch/papers/Berio-GI2017.pdf)
bib(http://calinon.ch/papers/Berio-GI2017.bib)
(Ref. for Bezier curves as MPC with viapoints)
[8] EPFL EE613 course "Machine Learning for Engineers"
url(http://calinon.ch/teaching.htm)
(Ref. for machine learning teaching material)
Installation prerequisite
PbDlib requires:
- glfw3 (lightweight and portable library for managing OpenGL contexts, windows and inputs)
- GLEW (The OpenGL Extension Wrangler Library)
- LAPACK (Linear Algebra PACKage)
- Armadillo (C++ library for linear algebra & scientific computing)
Instructions are given below.
ImGui (graphical user interface library for C++, https://github.com/ocornut/imgui) and gfx_ui (a minimal geometry editing UI, https://github.com/colormotor/gfx_ui), are also used and provided as part of this package. They are both distributed under the MIT license (see the docs/ folder).
Dependencies installation on Debian and Ubuntu
glfw3
sudo apt-get install libglfw3-devInstallation from source
If libglfw3-dev is not available on your system, you can install it manually with:
git clone https://github.com/glfw/glfw.git
cd glfw
mkdir build
cd build
cmake -DBUILD_SHARED_LIBS=ON ../
make
sudo make install
export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/usr/local/libNote: The -DBUILD_SHARED_LIBS is necessary otherwise cmake will create a
static library.
GLEW
sudo apt-get install libglew-devLAPACK
sudo apt-get install liblapack-devArmadillo
See http://arma.sourceforge.net/download.html for instructions.
Compilation (on Linux and MacOS)
cd pbdlib-cpp
mkdir build
cd build
cmake ..
makeCompilation (on Windows)
First, download all the necessary dependencies. They are (but Armadillo) provided as a binary package on their own homepage. We recommend to download and extract them all in the same folder.
glfw3: https://www.glfw.org/download.html
GLEW: http://glew.sourceforge.net/index.html
Armadillo: http://arma.sourceforge.net/download.html
Then use cmake-gui (http://cmake.org) to generate a solution for you version of Visual Studio. You'll need to manually indicate the path to the following dependencies when requested:
-
ARMADILLO_DIR: Path to Armadillo -
GLEW_DIR: Path to GLEW -
GLFW_DIR: Path to GLFW
Once the solution has been generated, open it (pbdlib_gui.sln in your build folder)
and compile as usual.
