Vortragsdetails

Development of 3D printed fingerprint for robotic hands and grippers

Background
Texture, and shape recognition through the implementation of haptic sensors are being widely explored in robotics.
There are various technologies haptic sensors use, such as vibrations, force feedback, air vortex rings, and ultrasound.
However, the implementation of this kind of sensors may be limited, especially in robotic hands and grippers, resulting in a reduced degree of freedom created to prevent damage to the haptic sensor.  

Topic introduction, and solution
This project aims to address designing a 3D printer fingerprint pattern to enhance the body-borne vibration signal of a RH8D Adult size Robot Hand.
The amplified vibration signal, generated by the fabricated 3D printer texture pattern, is recorded by a contact microphone.
Incorporating this strategy allows mounting sensors inside the robot's casing to reduce the mechanical constraint and noise background.

Methods
Strategically the RH8D has been designed with removable surface parts where the 3D printed texture pattern will be mounted.
To optimize the design of the surface pattern, a vibration analysis is carried out by reducing the robot finder-hand to a damper, spring-mass system.
Three experimental iterations analyze the sensor response to kinesthetic and tactile exploration.

Results
The result of the experimental iterations is a 3D printed surface capable to improve vibrations generated by a R8HD with a group of 100 different objects.
As an extra outcome, the data provided for the third experiment is collected as a haptic data-set which will be used to train a machine-learning algorithm to identify the physical properties.

In der Regel sind die Vorträge Teil von Lehrveranstaltungsreihen der Universität Bremen und nicht frei zugänglich. Bei Interesse wird um Rücksprache mit dem Sekretariat unter sek-ric(at)dfki.de gebeten.

© DFKI GmbH
last updated 30.07.2019
to top