A. René Geist

Postdoc, University of Tübingen

rene.geist [AT] uni-tuebingen.de

About

I'm a postdoc at the Tübingen AI Center focused on AI-driven robot simulation and control. Simulation is a cornerstone of modern robotics as many of today's most impressive real-world stunts - such as humanoid robots jumping over obstacles - are first learned in carefully designed virtual environments. The fidelity and structure of these simulators strongly shape what robots can ultimately achieve outside the lab.

My research develops high-capability simulations of the physical world by combining machine learning with computational physics. Current challenges I work on include differentiable contact dynamics, AI-driven particle simulations, narrowing the sim-to-real gap, efficient domain randomization, and modeling soft-body interactions, among others.

Publications

Most recent publications on Google Scholar.
A star besides an author's name indicates equal contribution.

SoftJAX and SoftTorch: Empowering Automatic Differentiation Libraries with Informative Gradients

SoftJAX and SoftTorch: Empowering Automatic Differentiation Libraries with Informative Gradients

Anselm Paulus*, A. René Geist*, Vít Musil, Sebastian Hoffmann, Georg Martius

February 2026

Learning 3D-Gaussian Simulators from RGB Videos

Learning 3D-Gaussian Simulators from RGB Videos

Mikel Zhobro, A. René Geist*, Georg Martius*

February 2026

arXiv
Hard Contacts with Soft Gradients: Refining Differentiable Simulators for Learning and Control

Hard Contacts with Soft Gradients: Refining Differentiable Simulators for Learning and Control

Anselm Paulus*, A. René Geist*, Pierre Schumacher, Vít Musil, Georg Martius

ICLR'26: International Conference on Learning Representations. 2026.

arXiv
Learning with 3D rotations, a hitchhiker's guide to SO(3)

Learning with 3D rotations, a hitchhiker's guide to SO(3)

A. René Geist, Jonas Frey, Mikel Zhobro, Anna Levina, Georg Martius

ICML'24: International Conference on Machine Learning. 2024.

Paper arXiv
Learning Deformable Linear Object Dynamics From a Single Trajectory

Learning Deformable Linear Object Dynamics From a Single Trajectory

Shamil Mamedov, A. René Geist, Ruan Viljoen, Sebastian Trimpe, Jan Swevers

IEEE Robotics and Automation Letters. 2025.

Paper arXiv
The Wheelbot: A jumping reaction wheel unicycle

The Wheelbot: A jumping reaction wheel unicycle

A. René Geist, Jonathan Fiene, Naomi Tashiro, Zheng Jia, Sebastian Trimpe

IEEE Robotics and Automation Letters. 2022.

Paper Video Code
SoftJAX and SoftTorch: Empowering Automatic Differentiation Libraries with Informative Gradients

SoftJAX and SoftTorch: Empowering Automatic Differentiation Libraries with Informative Gradients

Anselm Paulus*, A. René Geist*, Vít Musil, Sebastian Hoffmann, Georg Martius

February 2026

Smoothly Differentiable and Efficiently Vectorizable Contact Manifold Generation

Smoothly Differentiable and Efficiently Vectorizable Contact Manifold Generation

Onur Beker, A. René Geist, Anselm Paulus, Nico Gürtler, Ji Shi, Sylvain Calinon, Georg Martius

January 2026

Learning 3D-Gaussian Simulators from RGB Videos

Learning 3D-Gaussian Simulators from RGB Videos

Mikel Zhobro, A. René Geist*, Georg Martius*

February 2026

arXiv
Hard Contacts with Soft Gradients: Refining Differentiable Simulators for Learning and Control

Hard Contacts with Soft Gradients: Refining Differentiable Simulators for Learning and Control

Anselm Paulus*, A. René Geist*, Pierre Schumacher, Vít Musil, Georg Martius

ICLR'26: International Conference on Learning Representations. 2026.

arXiv
A Smooth Analytical Formulation of Collision Detection and Rigid Body Dynamics With Contact

A Smooth Analytical Formulation of Collision Detection and Rigid Body Dynamics With Contact

Onur Beker, Nico Gürtler, Ji Shi, A. René Geist, et al.

IROS'25: IEEE/RSJ International Conference on Intelligent Robots and Systems. 2025.

arXiv
Learning with 3D rotations, a hitchhiker's guide to SO(3)

Learning with 3D rotations, a hitchhiker's guide to SO(3)

A. René Geist, Jonas Frey, Mikel Zhobro, Anna Levina, Georg Martius

ICML'24: International Conference on Machine Learning. 2024.

Paper arXiv
Learning Deformable Linear Object Dynamics From a Single Trajectory

Learning Deformable Linear Object Dynamics From a Single Trajectory

Shamil Mamedov, A. René Geist, Ruan Viljoen, Sebastian Trimpe, Jan Swevers

IEEE Robotics and Automation Letters. 2025.

Paper arXiv
Pseudo-rigid body networks: learning interpretable deformable object dynamics from partial observations

Pseudo-rigid body networks: learning interpretable deformable object dynamics from partial observations

Shamil Mamedov*, A. René Geist*, Jan Swevers, Sebastian Trimpe

IROS'24: IEEE/RSJ International Conference on Intelligent Robots and Systems. 2024.

arXiv Code
The Wheelbot: A jumping reaction wheel unicycle

The Wheelbot: A jumping reaction wheel unicycle

A. René Geist, Jonathan Fiene, Naomi Tashiro, Zheng Jia, Sebastian Trimpe

IEEE Robotics and Automation Letters. 2022.

Paper Video Code
Physics-informed regression of implicitly-constrained robot dynamics

Physics-informed regression of implicitly-constrained robot dynamics

A. René Geist

PhD Thesis, University of Stuttgart. 2022.

Paper
Using physics knowledge for learning rigid-body forward dynamics with Gaussian process force priors

Using physics knowledge for learning rigid-body forward dynamics with Gaussian process force priors

Lucas Rath*, A. René Geist*, Sebastian Trimpe

CoRL'22: Conference on Robot Learning. 2022.

Paper
Structured learning of rigid-body dynamics: A survey and unified view from a robotics perspective

Structured learning of rigid-body dynamics: A survey and unified view from a robotics perspective

A. René Geist, Sebastian Trimpe

GAMM-Mitteilungen, Special Issue: Scientific Machine Learning. 2021.

Paper
Learning constrained dynamics with Gauss principle adhering Gaussian processes

Learning constrained dynamics with Gauss principle adhering Gaussian processes

A. René Geist, Sebastian Trimpe

L4DC'20: Conference on Learning for Dynamics and Control. 2020.

Paper
Learning environmental field exploration with computationally constrained underwater robots: Gaussian processes meet stochastic optimal control

Learning environmental field exploration with computationally constrained underwater robots: Gaussian processes meet stochastic optimal control

Daniel Andre Duecker*, Andreas René Geist*, Edwin Kreuzer, Eugen Solowjow

Sensors. 2019.

Paper
Data collection for robust end-to-end lateral vehicle control

Data collection for robust end-to-end lateral vehicle control

A. René Geist, Andreas Hansen, Eugen Solowjow, Shun Yang, Edwin Kreuzer

DSCC'17: ASME Dynamic Systems and Control Conference. 2017.

Paper Video
Embedded spherical localization for micro underwater vehicles based on attenuation of electro-magnetic carrier signals

Embedded spherical localization for micro underwater vehicles based on attenuation of electro-magnetic carrier signals

Daniel-André Duecker, A. René Geist, Michael Hengeler, Edwin Kreuzer, Marc-André Pick, Viktor Rausch, Eugen Solowjow

Sensors. 2017.

Paper Video
Towards a hyperbolic acoustic one-way localization system for underwater swarm robotics

Towards a hyperbolic acoustic one-way localization system for underwater swarm robotics

Andreas René Geist, Axel Hackbarth, Edwin Kreuzer, Viktor Rausch, Michael Sankur, Eugen Solowjow

ICRA'16: IEEE International Conference on Robotics and Automation. 2016.

Paper

Teaching

2025
Guest lecturer Reinforcement Learning
University of Tübingen · Georg Martius
Value estimation with Monte Carlo and TD learning
Teaching Assistant Reinforcement Learning
University of Tübingen · Georg Martius
2024
Guest lecturer Basics of Machine Learning
University of Tübingen · Georg Martius
Ridge regression, Bayesian regression, Decision trees, Random Forest
Guest lecturer Reinforcement Learning (Proseminar)
University of Tübingen · Georg Martius
From discrete to continuous deep reinforcement learning
Teaching Assistant Reinforcement Learning (Proseminar)
University of Tübingen · Georg Martius
2023
Teaching Assistant Computer Science in Mechanical Engineering
RWTH Aachen · Prof. Sebastian Trimpe
Guest lecturer Fundamentals of Machine Learning
RWTH Aachen · Prof. Sebastian Trimpe
Learning dynamics with probabilistic models
2022
Teaching Assistant Ethics of Artificial Intelligence and Robotics
RWTH Aachen · Dr. Dominik Baumann

Vita

Full CV (PDF)

I'm a postdoc working with Georg Martius at the University of Tübingen (until recently we've been located at the MPI-IS). During my PhD supervised by Sebastian Trimpe at the Max Planck Institute for Intelligent Systems, I worked on physics-informed machine learning of implicitly-constrained robot dynamics. In collaboration with Jonathan Fiene, we developed the "Wheelbot" being a robot platform for teaching control theory and robotics.

I worked on end-to-end vehicle control as a research intern at the Vehicle Dynamics and Control laboratory of the University of California, Berkeley. During my Master's in Theoretical Mechanical Engineering at TU Hamburg, I worked as a research student on autonomous field exploration via AUV swarms working with Eugen Solowjow.

A. René Geist
rene.geist [AT] uni-tuebingen.de