/Katharina Muelling

Katharina Muelling

Portrait of Katharina Muelling
Systems Scientist
Phone: (412) 692-1097
Personal Homepage
Administrative Assistant: Christine Downey

Mailing Address

My work is driven by the passion of getting robots out of research laboratories into the real world. To this end, I seek to understand how complex motor behavior can be modeled and learned, how motor skills can be generalized, how humans and robots can work together to achieve complex tasks, and how we can use these technologies to improve the quality of life for people with disabilities.

Robotic systems that are able to perform various tasks in human-inhabited and unstructured environments require robust movement generation and manipulation skills that compensate for uncertainties and disturbances in the environment. Such systems need to autonomously adapt to a highly dynamic environment while simultaneously accomplishing the task at hand. Hand-crafted solutions based on analytical engineering often fail to produce this form of adaptive behavior and are limited to the specific scenarios considered by the designer. Here, machine learning is a promising tool to create robotic systems that are able to adapt to a dynamic, unstructured environment. Most generic machine learning approaches how- ever, fail to learn on real robot platforms due to specific constraints and restrictions such as the real-time requirements and hardware exhaustion.

My work focuses on finding the limits of analytical engineering solutions and the development of robot learning algorithms that circumvent these limits. A fundamental problem for the development of robot learning methods is the necessity to achieve complex behaviors with a feasible amount of training data. Human demonstrations can be used to initialize robot learning approaches and reduce the learning time significantly. Furthermore, it provides a natural way for humans to teach robots and allows robots to acquire human-like behavior which is beneficial for human-robot interaction. Therefore, learning from and with humans is a central part of my work.

In the following, I will provide an overview of my research to date, together with research directions which I intend to pursue as a system faculty member at Carnegie Mellon University. The presented work is centered around (i) my doctoral studies, which focused on the development of a table tennis framework for an anthropomorphic robot arm and (ii) my current research that assists a tetraplegic person in teleoperating a robot arm to successfully perform everyday manipulation tasks such as opening a door.

Displaying 32 Publications

Current RI PhD Students