Modeliranje ponašanja heterogenog robotskog sustava probabilističkim metodama

master's thesis

• Title (english): Probabilistic planning for behavior-based heterogeneous robotic system
• Author: Nina Družetić
• Mentor: Matko Orsag (mentor)
• Committee member: Matko Orsag (predsjednik povjerenstva)
• Committee member: Stjepan Bogdan (član povjerenstva)
• Committee member: Zdenko Kovačić (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2018-07-12, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Electrical Engineering

Abstract

Cilj ovog rada je implementacija sustava modeliranja ponašanja heterogenog robotskog sustava zasnovanog na djelomično osmotrivim Markovljevim procesima odlučivanja (POMDP). Nakon ulaska u nepoznati zatvoreni prostor, heterogeni robotski sustav prima opažanja o stanju prostorija koristeći senzor zvuka te klasificira prijateljske i neprijateljske jedinice unutar istih. Nakon klasifikacija jedinica, provode se akcije izbjegavanja prostorije, ukoliko se radi o neprijateljskoj jedinici ili ulaska u prostoriju ukoliko je klasificirana prijateljska jedinica. Algoritam je implementiran koristeći Python i Robotski operacijski sustav(ROS), a simulacije su izvedene koristeći Gazebo. Simulacijski eksperimenti pokazuju vrlo dobro ponašanje sustava u različitim situacijama pa se može zaključiti da model daje zadovoljavajuće rezultate.

Abstract (english)

This thesis implements probabilistic planning for heterogeneous robotic system using partially observable Markov decision processes (POMDP). After entering an unknown closed space, heterogeneous robotic system uses sound sensor to get observations of room's state and classifies hostages from terrorists. If there are terrorists in the room, actions that avoid that room are made. In other cases, when there are hostages in the room, system takes actions to enter the room.
Algorithm was implemented using Python and Robot Operating System (ROS), and tested with Gazebo.
To conclude, simulations show promising results in different scenarios and model gives satisfactory results.

• Keywords: heterogeni robotski sustav; POMDP; ROS; Gazebo
• Keywords (english): heterogeneous robotic systems; POMDP; ROS; Gazebo
• Language: croatian
• URN:NBN: urn:nbn:hr:168:928621
• Study programme: Title: Electrical Engineering and Information Technology; Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra inženjer/inženjerka elektrotehnike i informacijske tehnologije (magistar/magistra inženjer/inženjerka elektrotehnike i informacijske tehnologije)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2019-03-13 20:02:33