Assessing and Validating Robot Apparent Masses in Collision

Institut

TUM Munich Institute of Robotics and Machine Intelligence (Institut)

Typ

Semesterarbeit / Masterarbeit /

Inhalt

experimentell / konstruktiv /  

Beschreibung

In the recent years robot autonomy is becoming more and more feasible. Nevertheless, the safety of humans around the systems has to be ensured. For this, robots need to understand themselves, which kinds of injuries they can cause.
Paramount to this, is for the robot to correctly assess their own power and energy during motion. This requires adequate modeling of the robot dynamics parameters and sesnitivity of the robot to collision events.

In this work, we build up on existing work and rework a pendulum test setup for assessing collision in unconstrained space, such that we can understand the effects of robot collision sensitivity, accurate dyanmics modelling and the resulting robot masses in collision on human injury. 

Voraussetzungen

- basic understanding of mechanical principles

- programming skills (LabView, python, C++)

- strong interest in real-world data-collection

- beneficial are first experiences with sensing systems, test setup design and robotics

Möglicher Beginn

sofort

Kontakt

Robin Jeanne Kirschner
robin-jeanne.kirschnertum.de