Reinhard M. Grassmann

I am a Ph.D. Candidate with the Department of Computer Science and a research assistant with the Continuum Robotics Laboratory at the University of Toronto, Canada. I work on continuum robotics and machine learning. Before all that, I completed a B.Sc. and a M.Sc. Degree in Mechatronics at the Leibniz University Hannover, Germany in 2016 and 2018, respectively. From 2018 to 2019 I worked as a research assistant at the Lehrstuhl fuer Kontinuumsrobotik at the Leibniz University Hannover.

Email  /  Bio  /  Google Scholar  /  GitHub  /  Semantic Scholar

I'm interested in continuum robotics, machine learning, and dual quaternions. Aside from that, creativity, innovation, and philosophy of science are also part of my field of interest to some extent.

During my research adventures, I have gratefully collaborated or am still collaborating with (in alphabetical order) J. Burgner-Kahrs, R. Z. Chen, P. T. Dewi, S. Haddadin, T. Hamoda, L. Johannsmeier, V. Modes, P. Le, N. Liang, S. Lilge, Q. Peyron, P. Rao, A. Senyk, and, C. Shentu.

We propose the Clarke transform and Clarke coordinates enabling the development of methods on the two-dimensional manifold embedded in the n-dimensional joint space.

We derive and investigate the lower triangular transformation matrix to disentangle the tube inequalities.

To democratize continuum robots research, we propose an actuation module to build torque-controlled continuum robots and provide open-source software and hardware with our initiative called the Open Continuum Robotics Project.

We provide the first public dataset for a three-tube CTCR to democratize research on learning-based and physics-based modelling of the kinematics.

To achieve variable tendon routing, we propose a segment design that combines two distinct characteristics of tendon-driven continuum robots, i.e. variable length and non-straight tendon routing, into a single segment by enabling rotation of its backbone.

We introduce methodologies to compute the inverse kinematics for concentric tube continuum robots from a desired shape as input.

We discuss the impact of the current habit of developing a variety of different prototypes on the emergence of a ubiquitous robotic platform.

We address the problem of generating a singularity-free trajectory for multiple via poses in SE(3), while complying with the requirement of C4 continuity.

We investigate the influence of different joint space and orientation representations on the approximation of the forward kinematics.

We introduce a point-to-point trajectory planner for serial robotic structures that combines the ability to avoid self-collisions and to respect motion constraints, while complying with the requirement of being C4 continuous.

We introduce a joint description to learn the forward and inverse kinematics for concentric tube continuum robots from real robot measurements.

Theses that I wrote during my mechatronics studies at Leibniz University Hannover.

"Artificial Neural Networks for Learning Forward and Inverse Kinematics of Tubular Continuum Robots"

Master’s thesis supervised by Prof. Dr.-Ing. Jessica Burgner-Kahrs

"Generating Smooth Trajectories for Obstacle Avoidance via Quaternions in Cartesian Space"

Bachelor’s thesis supervised by Prof. Dr.-Ing. Sami Haddadin

"Data Fusion Using an Extended Kalman Filter for an Optically Navigated Light Weight Robot"

Project thesis supervised by Prof. Dr.-Ing. Tobias Ortmaier

Research projects with my involvement.

"Open Continuum Robotics Project" (link to OpenCR Project, link to project description)

Principle investigator is Prof. Dr.-Ing. Jessica Burgner-Kahrs

Relevant papers:
- Open Continuum Robotics – One Actuation Module to Create them All
- CTCR Prototype Development: An Obstacle in the Research Community?

"Learning the Kinematics of Tubular Continuum Robots" (link to project description)

Principle investigator is Prof. Dr.-Ing. Jessica Burgner-Kahrs

Relevant papers:
- A Dataset and Benchmark for Learning the Kinematics of Concentric Tube Continuum Robots
- Learning-based Inverse Kinematics from Shape as Input for Concentric Tube Continuum Robots
- On the Merits of Joint Space and Orientation Representations in Learning the Forward Kinematics in SE(3)
- Learning the Forward and Inverse Kinematics of a 6-DOF Concentric Tube Continuum Robot in SE(3)

"Algorithms for Continuum Robots" (link to project description)

Principle investigator is Prof. Dr.-Ing. Jessica Burgner-Kahrs

Relevant papers:
- Clarke Transform — A Fundamental Tool for Continuum Robotics
- On the Disentanglement of Tube Inequalities in Concentric Tube Continuum Robots
- Quaternion-Based Smooth Trajectory Generator for Via Poses in SE(3) Considering Kinematic Limits in Cartesian Space

I made it in the acknowledgment section of the following papers :)