Bringing geometric methods to robot learning, optimization and control

October 29th, 2020

Thank you all, it was awesome!

Invited talks

The talks of the invited speakers are available in our Youtube playlist ! From October 25th, you can also watch them on the IROS on-demand platform.

Live session

When: October 29th, 2020, 13:00-16:00 CET (08:00-11:00 GMT-4, 21:00-00:00 GMT+9)

The recorded live session is now available online! Check it out here.

Call for Papers

Important dates

  • September 11th, 2020 (23:59:00 PST) : Submission deadline

  • September 30th, October 2nd 2020: Notification of acceptance

  • October 16th, 2020 (23:59:00 PST) : Final submission

  • October 25th, October 29th, 2020: Workshop live session

Submission website

Topics of interest

The topics of interest include (but are not limited to) the following:

  • Geometry-awareness

  • Riemannian manifolds

  • Lie Algebra

  • Robot learning and control

  • Optimization

  • Motion planning

  • Manifold learning

  • Robot modeling

  • Topological methods in robotics

Instructions for authors

We welcome prospective participants to submit an extended abstract (max. 2 pages) in PDF format and prepared using the standard IEEE template for IROS. Submissions will be made via Microsoft CMT at the following link: Each abstract will be peer-reviewed and selected based on their originality, relevance, technical clarity, and presentation. The accepted abstracts will be made available on the workshop's website (upon authors agreement). Accepted contributions will be presented during a poster session, whose format will follow the general format of IROS and will be open to online attendees in any case.

The Bosch Center for Artificial Intelligence (BCAI) will sponsor an award of 500 euros for the best contribution presented during our workshop!

Instructions for camera-ready version

We invite the authors of accepted submissions to upload a camery-ready packet, via CMT, including

  • the revised version of the extended abstract including reviewers' suggestions;

  • a spotlight video recording of 4-5 minutes long in MP4 format (min. 480 pixels height, 16:9 aspect ratio).

Outline and Objectives

In many robotic applications, robots are required to react to new situations, act in unstructured and dynamic environments, and overcome uncertainty. This entails to have outstanding adaptation capabilities so that robot actions lead to successful performance. A key component in both data-driven learning and adaptation is how robots may exploit explicit (e.g. domain knowledge) or implicit (e.g. learned) structures arising in the collected data. Domain knowledge and data structures in robotics can be viewed from a geometric perspective as different variables and problems have specific geometric characteristics. Rigid body orientations, controller gains, inertia matrices, manipulability ellipsoids or end-effector poses are examples of variables with predefined geometric structure. These diverse types of variables do not belong to a vector space and thus the use of classical Euclidean space methods for treating and analyzing these variables is inadequate. In this context, differential geometry, or more specifically Lie group and Riemannian manifold theories provide appropriate tools and methods to cope with or to learn the geometry of non-Euclidean parameter spaces.

The main objective of this workshop is to attract the interest of the robotic community on geometric methods, which have been overlooked in robot learning, control and optimization. Moreover, we expect this workshop will raise awareness on the importance of geometry in the different research branches of robotics. We aim at bringing together researchers from various robotic fields to discuss the benefits and explore the challenges of bringing geometry-awareness to solve robotic problems. Finally, we aim at building bridges between the robotic community and mathematicians, as well as machine learning researchers, in order to efficiently tackle the upcoming challenges involving differential geometry and robotics.


3-hour live session

October 29th, 2020, 13:00-16:00 CET (08:00-11:00 GMT-4, 21:00-00:00 GMT+9)

Time (GMT+2)

13:00 - 13:10

13:10 - 13:35

13:35 - 14:15

14:15 - 14:30

14:30 - 15:00

15:00 - 15:35

15:35 - 16:00


Introduction by the organizers

Q&A session for the talks of L. Carlone, J. Solà, T. Lee and B. Adorno

Poster session part 1 (accepted contributions 1-7, see order below)

Coffee break

Q&A session for the talks of S. Bhattacharya, M. Likhachev, A. Varava, M. Mukadam and L. Malagò

Poster session part 2 (accepted contributions 8-13, see order below)

Panel discussion

Invited Speakers

Massachusetts Institute of Technology

Certifiable 3D Perception: from Geometry to Global Optimization and back

Institut de Robòtica I Informàtica Industrial

Micro Lie theory: a crash course on Lie theory for the roboticist with applications to motion estimation, SLAM and sensor self-calibration

Seoul National University

Geometric methods for dynamic model-based robotics: identification, adaptive control and excitation trajectory optimization

Federal University of Minas Gerais

Complex Robotic Systems: Modeling, Control, and Planning using Dual Quaternion Algebra

Planning and Control on Riemannian Manifolds with Boundaries

Carnegie Mellon University

Planning for High-dimensional Robotic Systems by Solving Problems in Low-dimensional Manifolds

KTH Royal Institute of Technology

State space representations for complex manipulation: integrating geometry, topology, and machine learning

Structured Policies for Reactive Motion Generation

Romanian Institute of Science and Technology

Lagrangian and Hamiltonian Mechanics for Probabilities on the Statistical Manifold

Accepted contributions

  1. Meng Guo, Lukas Schwenkel, and Mathias Buerger. Optimizing Sequences of Probabilistic Manipulation Skills Learned from Demonstration.

  2. Hadi Beik-mohammadi, Leonel Rozo, Gerhard Neumann, and Søren Hauberg. Riemannian Manifold Learning for Robot Motion Skills.

  3. Sylvain Chevallier, Marie-Constance Corsi, Florian Yger, and Camille Noûs. Extending Riemannian Brain-Computer Interface to Functional Connectivity Estimators.

  4. Filip Marić, Matthew Giamou, Ivan Petrović, and Jonathan Kelly. Inverse Kinematics as Low-Rank Euclidean Distance Matrix Completion.

  5. Maxime A. Vaidis, Johann Laconte, Vladimir Kubelka, and Francois Pomerleau. Improving the Iterative Closest Point Algorithm using Lie Algebra.

  6. Kira Ruschmeier, Philipp Schillinger, and Leonel Rozo. CMAES on Riemannian Manifolds for Optimizing Robotic Manipulation Tasks.

  7. Lander Vanroye, Maxim Vochten, Erwin Aertbeliën, Wilm Decré, and Joris De Schutter. Efficient Optimization-based Calculation of Coordinate-Invariant Trajectory Shape Descriptors for on-line applications.

  8. Xuesu Xiao, Bo Liu, and Peter Stone. Motion Planners Learned from Geometric Hallucination.

  9. Moonyoung Lee, Youngsun Kwon, Sebin Lee, Hyobin Jeong, Yujin Heo, Minsu Kim, and Jun-Ho Oh. Geometric Footstep Planning and Control for Dynamic Humanoid Locomotion over Uneven Terrain.

  10. Muhammad Suhail Saleem, Raghav Sood, and Maxim Likhachev. Topology Guided Path Planning for a Snake Robot in Cluttered Environments.

  11. Brian E. Jackson, and Zachary Manchester. Planning With Attitude.

  12. Vladislav Polianskii, Anastasiia Varava, Danica Kragic, and Florian T. Pokorny. Configuration space approximations in higher dimensions.

  13. Michael Psenka, Tolga Birdal, and Leonidas Guibas. Reconstruction without Registration.

Best workshop paper award, sponsored by the Bosch Center for Artificial Intelligence

Filip Marić, Matthew Giamou, Ivan Petrović, and Jonathan Kelly. Inverse Kinematics as Low-Rank Euclidean Distance Matrix Completion.


Noémie Jaquier

Karlsruhe Institute of Technology

Leonel Rozo

Bosch Center for Artificial Intelligence

Søren Hauberg

Technical University of Denmark

Hans-Peter Schröcker

Universität Innsbruck

Suvrit Sra

Massachusetts Institute of Technology


This workshop is sponsored by the Bosch Center for Artificial Intelligence (BCAI). BCAI sponsored an award for the best contribution presented during the workshop poster session.