Predicting Robot Trajectory Deviations using Machine Learning

Institute

Lehrstuhl für Angewandte Mechanik (TUM-ED)

Type

Semester Thesis / Master's Thesis /

Content

experimental / theoretical /  

Description

Topic:

All robots deviate from the trajectory that’s planned for them. The amount and type of deviation varies based on the design and control of the robot. Causes may include unexpected flexibility in the links, motor cogging, or transmission dynamics among other things. For some high-precission applications this error is not tolerable and must be compensated. However before it can be compensated it must be predicted so that the robot plan can be adjusted before error occurs. This thesis will use a data-driven approach to predict the robot error for a given trajectory. To train the neural network, trajectory measurement campaigns at various speeds and locations in the robot space will be conducted. Different network architectures and hyper-parameters will be evaluated to find the optimal predictor. Robustness of the network against noisy or incomplete datasets will be tested. The ability to extrapolate beyond the trained data-set will also be assessed.

 

As with all my theses, the proposed topic is just a starting point. It is flexible and can be adjusted to what interests you and what direction the research leads. The expectations regarding results, prerequisits and work-ethic will be modulated based on whether a MA/SA or BA is written.

 

Application:

Please send all previous transcripts of records and your full CV to tomas.slimak@tum.de with the subject Application MA/SA/BA "name of thesis". Write a few sentences about what motivates you to apply for this thesis and why you would fit the topic. Please be specific in citing the experience/expertise that you have which would be relevant to this topic.

 

After submitting your application, it will be reviewed and if you are deemed suitable for the topic, you will be invited for an interview. Please be aware that part of this interview will be an oral evaluation of your background and understanding of concepts relevant to the thesis. The more preparation, creativity and initiative you are able to show, the better.

 

If you are interested in multiple theses that I am offering, do not send multiple applications, just name all the titles in a single mail. Applications and theses can be submitted in English or German as well.

 

Requirements

 

Possible start

sofort

Contact

Tomas Slimak, M.Sc.
Room: 3104
Phone: +49 (89) 289 - 15226
tomas.slimaktum.de