Upravljanje robota interpretacijom ljudskih kretnji

undergraduate thesis

• Title (english): Robot control by interpretation of human gestures
• Author: Bruno Dobrić
• Mentor: Bojan Jerbić (mentor)
• Granter: University of Zagreb; Faculty of Mechanical Engineering and Naval Architecture; Zagreb
• Defense date and country: 2018-09-27, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Mechanical Engineering

Abstract

Interakcija čovjeka i robota predmet je brojnih istraživanja još od vremena prije oblikovanja robotike kao znanosti. Mnogi aspekti tih istraživanja temelje se na proučavanju komunikacije između ljudi nastojeći odnos čovjeka i robota dovesti na jednako tako prirodnu i intuitivnu razinu. Budući da su roboti u suvremenom dobu sastavni dio svakodnevnog okruženja različitog od onog uređenog industrijskog, javlja se potreba za razvojem kognitivnih upravljačkih modela koji bi omogućili ljudima interaktivno upravljanje, a robotima donijeli mogućnost razumijevanja svoje okoline. Ovaj završni rad prikazuje postupak razvoja programske podrške koja omogućuje upravljanje robotom interpretacijom ljudskih kretnji. Prikazani postupak obuhvaća izradu aplikacije koja pomoću stereo-vizijskog sustava Microsoft Kinect snima ljudske pokrete, razvoj klasifikacijskih algoritama za prepoznavanje pokretu u programskom paketu Matlab te izradu poslužiteljske aplikacije koja omogućuje komunikaciju s robotom. Na ovaj način omogućeno je intuitivno upravljanje robotom bez potrebe znanja programiranja, nalik neverbalnoj ljudskoj komunikaciji.

Abstract (english)

Human-robot interaction has been a subject of numerous researches before robotics per se. Many aspects of these researches are based on the study of communication between people trying to bring the human-robot interaction to an equally natural and intuitive level. Nowadays robots are an integral part of a daily environment – the one that differs from those found in sophisticated industry. Because of that, the need for developing models of cognitive control that would enable people to interactively manage and provide robots with an understanding of their environment grows daily. This paper demonstrates the process of developing a program that enables robot control by interpretation of human gestures. The process described includes the creation of an application that uses the Microsoft Kinect Stereo Vision system to capture human movements, development of algorithms for gesture recognition in Matlab software package and creation of a server application that enables communication with the robot. This way intuitive robot control is enabled without the need for advanced programming skills.

• Keywords: interakcija čovjeka i robota; umjetna inteligencija; strojno učenje
• Keywords (english): Human-robot interaction; artificial intelligence; machine learning
• Language: croatian
• URN:NBN: urn:nbn:hr:235:885986
• Study programme: Title: Mechanical Engineering; Study programme type: university; Study level: undergraduate; Academic / professional title: sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka strojarstva (sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka strojarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Open access
• Created on: 2019-05-22 23:03:14