Header logo is


2017


Thumb xl fig toyex lqr1kernel 1
On the Design of LQR Kernels for Efficient Controller Learning

Marco, A., Hennig, P., Schaal, S., Trimpe, S.

Proceedings of the 56th IEEE Annual Conference on Decision and Control (CDC), pages: 5193-5200, IEEE, IEEE Conference on Decision and Control, December 2017 (conference)

Abstract
Finding optimal feedback controllers for nonlinear dynamic systems from data is hard. Recently, Bayesian optimization (BO) has been proposed as a powerful framework for direct controller tuning from experimental trials. For selecting the next query point and finding the global optimum, BO relies on a probabilistic description of the latent objective function, typically a Gaussian process (GP). As is shown herein, GPs with a common kernel choice can, however, lead to poor learning outcomes on standard quadratic control problems. For a first-order system, we construct two kernels that specifically leverage the structure of the well-known Linear Quadratic Regulator (LQR), yet retain the flexibility of Bayesian nonparametric learning. Simulations of uncertain linear and nonlinear systems demonstrate that the LQR kernels yield superior learning performance.

am ics pn

arXiv PDF On the Design of LQR Kernels for Efficient Controller Learning - CDC presentation DOI Project Page [BibTex]

2017


arXiv PDF On the Design of LQR Kernels for Efficient Controller Learning - CDC presentation DOI Project Page [BibTex]


Thumb xl teaser
Optimizing Long-term Predictions for Model-based Policy Search

Doerr, A., Daniel, C., Nguyen-Tuong, D., Marco, A., Schaal, S., Toussaint, M., Trimpe, S.

Proceedings of 1st Annual Conference on Robot Learning (CoRL), 78, pages: 227-238, (Editors: Sergey Levine and Vincent Vanhoucke and Ken Goldberg), 1st Annual Conference on Robot Learning, November 2017 (conference)

Abstract
We propose a novel long-term optimization criterion to improve the robustness of model-based reinforcement learning in real-world scenarios. Learning a dynamics model to derive a solution promises much greater data-efficiency and reusability compared to model-free alternatives. In practice, however, modelbased RL suffers from various imperfections such as noisy input and output data, delays and unmeasured (latent) states. To achieve higher resilience against such effects, we propose to optimize a generative long-term prediction model directly with respect to the likelihood of observed trajectories as opposed to the common approach of optimizing a dynamics model for one-step-ahead predictions. We evaluate the proposed method on several artificial and real-world benchmark problems and compare it to PILCO, a model-based RL framework, in experiments on a manipulation robot. The results show that the proposed method is competitive compared to state-of-the-art model learning methods. In contrast to these more involved models, our model can directly be employed for policy search and outperforms a baseline method in the robot experiment.

am ics

PDF Project Page [BibTex]

PDF Project Page [BibTex]


Thumb xl apollo system2 croped
Model-Based Policy Search for Automatic Tuning of Multivariate PID Controllers

Doerr, A., Nguyen-Tuong, D., Marco, A., Schaal, S., Trimpe, S.

In Proceedings of the IEEE International Conference on Robotics and Automation (ICRA), pages: 5295-5301, IEEE, Piscataway, NJ, USA, IEEE International Conference on Robotics and Automation (ICRA), May 2017 (inproceedings)

am ics

PDF arXiv DOI Project Page [BibTex]

PDF arXiv DOI Project Page [BibTex]


Thumb xl this one
Virtual vs. Real: Trading Off Simulations and Physical Experiments in Reinforcement Learning with Bayesian Optimization

Marco, A., Berkenkamp, F., Hennig, P., Schoellig, A. P., Krause, A., Schaal, S., Trimpe, S.

In Proceedings of the IEEE International Conference on Robotics and Automation (ICRA), pages: 1557-1563, IEEE, Piscataway, NJ, USA, IEEE International Conference on Robotics and Automation (ICRA), May 2017 (inproceedings)

am ics pn

PDF arXiv ICRA 2017 Spotlight presentation Virtual vs. Real - Video explanation DOI Project Page [BibTex]

PDF arXiv ICRA 2017 Spotlight presentation Virtual vs. Real - Video explanation DOI Project Page [BibTex]

2001


no image
Computational micromagnetism of magnetic structures and magnetization processes in thin plantelets and small particles

Kronmüller, H., Hertel, R.

In Magnetic Storage Sstems Beyond 2000, 41, pages: 345-362, Nato Science Series II: Mathematics, Physics and Chemistry, Kluwer Academic Publishers, Rhodos, Greece, 2001 (inproceedings)

mms

[BibTex]

2001


[BibTex]


no image
Hydrogen storage in mechanically treated single wall carbon nanotrubes

Haluska, M., Hulman, M., Hirscher, M., Becher, M., Roth, S., Stepanek, I., Bernier, P.

In Electronic Properties of Molecular Nanostructures: XV International Winterschool/Euroconference, 591, pages: 603-608, American Institute of Physics Conference Proceedings, AIP, Kirchberg [Austria], 2001 (inproceedings)

mms

[BibTex]

[BibTex]


no image
Isotopic mass and lattice constant of Si and Ge: X-Ray standing wave measurements

Zegenhagen, J., Kazimirov, A., Cao, L. X., Konuma, M., Sozontov, E., Plachke, D., Carstanjen, H. D., Bilger, G., Haller, E., Kohn, V., Cardona, M.

In Proceedings of the 25th Conference on the Physics of Semiconductors, 87, pages: 125-127, Springer proceedings in physics, Springer, Osaka, Japan, 2001 (inproceedings)

mms

[BibTex]

[BibTex]


no image
Positron Annihilation Studies on Stable and Undercooled Metal Melts at the Stuttgart Pelletron

Stoll, H., Siegle, A., Major, J.

In Application of Accelerators in Research and Industry, 576, pages: 749-752, AIP Conference Proceedings, Denton, Texas, USA, 2001 (inproceedings)

mms

[BibTex]

[BibTex]


no image
Submicrometer spatially resolved measurements of mechanical properties and correlation to microstructure and composition

Kunert, M., Baretzky, B., Baker, S. P., Mittemeijer, E. J.

In Fundamentals of Nanoindentation and Nanotribology II, 649, pages: Q3.2.1-Q3.2.6, Materials Research Society Symposium Proceedings, MRS, Boston, MA, USA, 2001 (inproceedings)

mms

[BibTex]

[BibTex]


no image
The six-jump diffusion cycles in B2-compounds

Drautz, R., Meyer, B., Fähnle, M.

In Proceedings of DIMAT 2000, the Fifth International Conference on Diffusion in Materials, pages: 417-422, Defect and Diffusion Forum, Scitec Publications Ltd., Paris, France, 2001 (inproceedings)

mms

[BibTex]

[BibTex]


no image
Ionic nitriding of austenitic and ferritic steel with the aid of a high aperture hall current accelerator

Straumal, B. B., Vershinin, N. F., Friesel, M., Ishenko, S. A., Gust, W.

In Diffusion in Materials DIMAT2000, 194, pages: 1457-1462, Defect and Diffusion Forum, Trans Tech, Paris, France, 2001 (inproceedings)

mms

[BibTex]

[BibTex]


no image
First proof of slow trapping of positronium in polymers by an Age-Momentum-Correlation (AMOC) experiment

Dauwe, C., Balcaen, N., van Waeyenberge, B., van Petegem, S., Stoll, H.

In Positron Annihilation. Proceedings of the 12th International Conference on Positron Annihilation, 363/365, pages: 254-256, Materials Science Forum, Trans Tech Publications Ltd., München, 2001 (inproceedings)

mms

[BibTex]

[BibTex]


no image
Positron-age-momentum correlation

Stoll, H., Bandzuch, P., Siegle, A.

In Positron Annihilation: Proceedings of the 12th International Conference on Positron Annihilation, 363-365, pages: 547-551, Materials Science Forum, Trans Tech Publications Ltd., München, 2001 (inproceedings)

mms

[BibTex]

[BibTex]


no image
Nanocrystalline and nanostructured high-performance permanent magnets

Goll, D., Hadjipanayis, G. C., Kronmüller, H.

In Applications of Ferromagnetic and Optical Materials, Storage and Magnetoelectronics, 674, pages: U2.4.1-U2.4.12, Materials Research Society Symposium Proceedings, MRS, San Francisco, Calif., 2001 (inproceedings)

mms

[BibTex]

[BibTex]


no image
Ion beam analysis with monolayer depth resolution using the electrostatic spectrometer at the MPI Stuttgart

Plachke, D., Blohm, G., Fischer, T., Khellaf, A., Kruse, O., Stoll, H., Carstanjen, H. D.

In Proceedings of the 16th International Conference on Applications of Accelerators in Research and Industry, 576, pages: 458-462, American Institute of Physics Conference Proceedings, AIP, Denton, Texas, 2001 (inproceedings)

mms

[BibTex]

[BibTex]


no image
From the electronic structure to the macroscopiy behavior: A multi-scale analysis of plasticity in intermetallic compounds

Fähnle, M., Kohlhammer, S., Bester, G.

In Influences of Interface and Dislocation Behavior on Microstructure Evolution, 652, pages: Y4.5.1.-Y4.5.12, Materials Research Society Symposium Proceedings, MRS, Boston, Mass., USA, 2001 (inproceedings)

mms

[BibTex]

[BibTex]


no image
Influence of the microstructure on the magnetic properties of giant-magnetostrictive TbDyFe films

Hirscher, M., Winzek, B., Fischer, S. F., Kronmüller, H.

In Smart Materials. Proceedings of the 1st Caesarium, pages: 23-37, Springer, Bonn, 2001 (inproceedings)

mms

[BibTex]

[BibTex]


no image
Materials analysis with monolayer depth resolution using MeV ion beams

Carstanjen, H. D.

In 117, Las Vegas, USA, 2001 (inproceedings)

mms

[BibTex]

[BibTex]


no image
Flux-line pinning in low-angle grain boundaries.

Albrecht, J., Leonhardt, S., Kronmüller, H.

In Proceedings 10th International Workshop on Critical Currents (IWCC 2001), pages: 41-43, Göttingen, Germany, 2001 (inproceedings)

mms

[BibTex]

[BibTex]


no image
Measurement of the low-temperature self-diffusivity of lithium by elastic recoil detection analysis

Wieland, O., Carstanjen, H. D.

In Proceedings of DIMAT 2000, the Fifth International Conference on Diffusion in Materials, 194/199, pages: 35-41, Defect and Diffusion Forum, Scitec Publications Ltd., Paris, France, 2001 (inproceedings)

mms

[BibTex]

[BibTex]


no image
From the electronic structure to the macroscopic behaviour: a multi-scale analysis of plasticity in intermetallic compounds

Fähnle, M., Kohlhammer, S., Bester, G.

In Influences of Interface and Dislocation Behavior on Microstructure Evolution, 652, pages: Y.4.5.1-Y.4.5.12, Materials Research Society Symposium Proceedings, MRS, Boston, Mass., 2001 (inproceedings)

mms

[BibTex]

[BibTex]


no image
Enhancement of the critical current density of YBa2Cu3O7-8-films by substracte irradiation

Leonhardt, S., Albrecht, J., Warthmann, R., Kronmüller, H.

In High-Tc Superconductors and Related Applications: Materials Science, Fundamental Properties, and Some Future Electronic Applications. Proceedings of the NATO Advanced Study Institute, 86, pages: 529-534, NATO Science Series 3. High Technology, Kluwer Academic Publishers, Albena, Bulgaria, 2001 (inproceedings)

mms

[BibTex]

[BibTex]


no image
AMOC studies of positronium in fine MgO powder

van Waeyenberge, B., Dauwe, C., Stoll, H.

In Positron Annihilation. Proceedings of the 12th International Conference on Positron Annihilation, 363/365, pages: 401-403, Materials Science Forum, Trans Tech Publications Ltd., München, 2001 (inproceedings)

mms

[BibTex]

[BibTex]


no image
Atomic defects and electronic structure of B2-FeAl, CoAl and NiAl

Fähnle, M., Meyer, B., Bester, G., Majer, J., Börnsen, N.

In Proceedings of DIMAT 2000, the Fifth International Conference on Diffusion in Materials, 194/199, pages: 279-285, Defect and Diffusion Forum, Scitec Publications Ltd., Paris, France, 2001 (inproceedings)

mms

[BibTex]

[BibTex]