Ph.D. Max Spahn
Delft, Netherlands
About
Driven by a passion for integrating robots into human environments, my career has centered on reactive motion planning. Over the past five years in academic research, I have pioneered novel methods for online motion planning in mobile manipulation, utilizing non-linear optimization and Riemannian geometry. Committed to the open-source community, I ensure my research and software are publicly accessible. Understanding the complexities and challenges of reliable robotics, I have actively participated in robotic competitions, and I am enthusiastic about joining a competitive team in my next role.
Skills
optimization: non-linear constrained optimization for dynamic programming, optimization on Riemannian manifolds, Bayesian optimization for hyperparameter tuning
robotics: forward and inverse kinematics, impedance control for compliant trajectory generation, receding horizon motion control, collision avoidance in human-shared environments in real-time
programming: ROS (noetic, melodic), python, git, C++, LaTeX, bash
project management: leading and contributing to collaborative research projects, supervising several undergraduate students
languages: German, English and French
Experience
2024-2025
Post-Doctoral Researcher @ Delft University of Technology
- Integrating non-Riemannian trajectory generation into manipulation tasks
- Leading a group of researchers to develop a live demo of
multiple mobile manipulators in a human-shared environment
2020-2024
Ph.D. in Robotics @ Delft University of Technology
- Thesis: Trajectory Generation beyond Model Predictive Control (Defense November 2024)
- Reactive trajectory generation for high-dimensional robots
- Using the geometry of the robot’s configuration space to generate consistent
- Combining several research projects into a proof-of-concept demonstration in supermarket automation
Education
2013-2019
RWTH Aachen, Mechanical Engineering, B.Sc. and M.Sc.
- focus on computational fluid dynamics and robotics
- Master thesis: Modeling High Weissenberg number flows in OpenFOAM
- Bachelor thesis: A Robust and Predictable Algorithm for the Numerical Solution of the Inverse Kinematics Problem for Serial Robots
2015-2017
École Centrale Supéléc Paris, M.Sc. (Double Degree T.I.M.E.)
- focus on fundamental mathematics
- several courses on algorithms and computer science
Publications
Journals
2023
M. Spahn, M. Wisse, J. Alonso-Mora: “Dynamic optimization fabrics for motion generation”, IEEE Transactions on Robotics
2022
C. Meo, G. Franzese, C. Pezzato, M. Spahn, P. Lanillos: “Adaptation through prediction: multisensory active inference torque control”, IEEE Transactions on Cognitive and Developmental Systems
2020
M. Frings, N. Hosters, C. Müller, M. Spahn, C. Susen, K. Key, S. Elgeti: SplinLib: A modern multipurpose C++ spline library”, Advances in Engineering Software
Conferences
2024
M. Spahn, C. Pezzato, C. Salmi, R. Dekker, C. Wang, C. Pek, J. Kober, J. Alonso-Mora, C. Hern'andez Corbato, M. Wisse: “Demonstrating Adaptive Mobile Manipulation in Retail Environments”, Robotics: Science and Systems
2023
M. Spahn, J. Alonso-Mora: “Autotuning Symbolic Optimization Fabrics for Trajectory Generation”, IEEE International Conference on Robotics and Automation
2023
S. Bakker, L. Knödler, M. Spahn, W. Böhmer, J. Alonso-Mora: “Multi-Robot Local Motion Planning Using Dynamic Optimization Fabrics”, IEEE International Symposium on Multi-Robot & Multi-Agent Systems
2022
M. Spahn, C. Salmi, J. Alonso-Mora: “Local Planner Bench: Benchmarking for Local Motion Planning”, Workshop at IEEE/RSJ International Conference on Intelligent Robots and Systems
2021
M. Spahn, B. Brito, J. Alonso-Mora: “Coupled Mobile Manipulation via Trajectory Optimization with Free Space Decomposition”, IEEE International Conference on Robotics and Automation
Projects
Geometric Fabrics for robot trajectory generation
- Geometric fabrics formulate trajectory generation as second order dynamical systems on (non-)Riemannian manifolds
- This implementation is robot-agnostic and has been used in several research projects
- The project uses continuous integration and modern software engineering practices
- https://github.com/tud-amr/fabrics
Generic robotics gymnasium wrapper for pybullet and MuJoCo
- generic robotics gymnasium wrapper for pybullet and MuJoCo
- the wrapper is used by several researchers and students
- https://github.com/maxspahn/gym_envs_urdf
Robothon 2023, Platonics Delft
- participated in the Robothon 2023 with the Platonics Delft Team
- completed all manipulation tasks using learning-from-demonstration
- https://platonics-delft.github.io
Interests
- learning from my child, running, football, vegan cooking