Skip to content Skip to navigation
See our Campus Ready site for the most up to date information about instruction.Campus ReadyCOVID Help
Mechatronics, Embedded Systems and Automation

MESA LAB welcomes German visitor: Tobias Fromm

April 17, 2013

MESA LAB welcomes Mr. Tobias Fromm, Ph.D. student, Robotics Group, Jacobs University Bremen, Germany.

His MEAM Graduate Seminar info (4/19/2013, 2-3PM, Friday, room SE1-138)

Title: Intelligent Robots For The Win: Selected Topics in Applied Machine Learning

 Abstract:  For various robotics applications, conventional programming has turned out to be a dead end. Especially when it comes to increasing the usability of robotics in everyday situations, the solutions to new challenges need to be learned by the robot in an autonomous way.

This talk focuses on three different, but coherent real-world applications of Machine Learning in the context of service and industrial robotics:

 1.) Low-level Learning from Demonstration in order to learn motoric tasks, e.g. pick-and-place,

 2.) High-level Learning from Demonstration in order to learn abstract goals, e.g. setting a table, and

 3.) Hierarchical Reinforcement Learning in order to learn dexterous manipulation tasks, e.g. turning a sphere with robotic fingers.

 All these topics will be introduced based on real-world examples of service robots and industrial prototypes.


 Speaker: Mr. Tobias Fromm, Ph.D. student, Robotics Group, Jacobs University Bremen, Germany

 About the Speaker:  Mr. Tobias Fromm is currently a Ph.D. student in the Robotics Group at Jacobs University Bremen, Germany. Tobi obtained his Master of Science degree in Computer Science at Ravensburg-Weingarten University of Applied Sciences. He spent 6 months at the Center for Self-Organizing and Intelligent Systems (CSOIS) at Utah State University for his MS thesis under supervision of Professor YangQuan Chen on real-time vision-based guidance for Unmanned Aerial Systems. His research interests include image processing and behavior planning in robotics with the help of Machine Learning.