Percepcija objekta pomoću senzora sile dodira

undergraduate thesis

• Title (english): Tactile perception using soft force sensor
• Author: Ivona Krajačić
• Mentor: Matko Orsag (mentor)
• Committee member: Matko Orsag (predsjednik povjerenstva)
• Committee member: Marsela Polić (član povjerenstva)
• Committee member: Antun Ivanović (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2018-07-13, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Electrical Engineering

Abstract

Glavna komponenta ovog rada je prikaz algoritma detekcije ruba objekta temeljenog na Bayesovoj analizi kao jedan od ključnih načela prepoznavanja objekta. Osnovno sredstvo s kojim se postiže realizacija ovog algoritma je senzor dodira TacTip, čije su osnovne komponente i način rada također objašnjeni. TacTip omogućuje dobivanje reprezentne slika, za svaki kut i radijalan pomak r, koji opisuju odnos senzora i ruba objekta, a čija kombinacija definira modele za razvoj i testiranje algoritma detekcije ruba. Slike modela dobivene TacTip-om, neuronskom mrežom su pretvorene u jedinstvene numeričke vrijednosti,koje su pogodne za rad sa
razvijenim Python skriptama. Ovakvim nalčinom rada, nepoznat ulaz seta podataka klasificira se u jedan od poznatih modela, tj. kombinacija kuta fi i pomaka r.

Abstract (english)

As a key principle in object detection, a Bayesian analysis based edge detection
algorithm is in the focus of this thesis.
The edge detection algorithm relies on the data provided by a tactile sensor TacTip,
whose basic components and functionality are described in the thesis.
TacTip provides sensory data for each combination of an angle theta and radial
displacement r, which describe spatial relation of the sensor and object edge. Each
combination of the angle and the displacement is considered a distinct class in edge
detection classification problem.
The sensory data acquired in the form of an image is encoded using a neural
network into a feature array of lower dimensionality.
Using the developed method, an unknown sensory reading can be classified into
one of the models described by angle theta and displacement r.

• Keywords: Detekcija ruba; Bayes; senzor dodira; TacTip
• Keywords (english): Object detection; Bayes; tactile sensor
• Language: croatian
• URN:NBN: urn:nbn:hr:168:079678
• Study programme: Title: Electrical Engineering and Information Technology; Study programme type: university; Study level: undergraduate; Academic / professional title: sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka elektrotehnike i informacijske tehnologije (sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka elektrotehnike i informacijske tehnologije)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2019-03-13 21:53:59