The essence of this master thesis is design and implementation of sensor system based on robotic framework which is called ROS (Robot Operating System). The main task is to perform detailed analysis and test of capabilities of the framework with final implementation on specific robot application (sensor system) with following evaluation of applicability of the system in mobile robotics. As parallel aim is to create detailed general and practical guide for beginners with ROS which they are also beginners in Linux based operating systems.

Through series of experiments this work compares effects of different types of genetic algorithms on evolution of a neural network that is used to control a robot. Genetic algorithms using binary and real coded individuals, algorithms using basic and advanced mutations and crossovers and algorithms using fixed and variable population size are compared on three tasks of evoltionary robotics. The goal is to determine wether usage of advanced genetic algorithms leads to faster convergence or to better solution than usage of basic genetic algorithm. Experiments are performed in an easily extendable simulator developed for purposes of this work.

Magistrinio darbo tikslas – įgyvendinti, graužikų biologinio, aplinkos žymėjimo kvapais ir navigacijos, proceso modelį robotikos srityje. Graužikai aplinkoje orientuojasi analizuodami informaciją gautą iš skirtingų penkių jutimo organų. Regėjimas padeda suvokti aplinkos erdvę, reljefą, atstumus iki objektų. Klausa, suteikia informaciją apie aplink esančius garso šaltinius, besiartinančius gyvūnus. Lytėjimas, padeda nustatyti objekto paviršiaus savybes: karštas ar šaltas, drėgnas ar sausas, minkštas ar kietas. Uoslė ir skonio receptoriai naudojami nustatyti maisto šaltiniui ar teritorijos riboms. Realiame biologiniame pasaulyje gyvūnai gali lengvai analizuoti informaciją gautą iš skirtingų jutimo organų ir ją derinti erdvinės navigacijos procese. Dirbtinėje sistemoje šie duomenys yra gaunami iš sensorių pasižyminčių skirtingomis savybėmis ir techniniais parametrais. Projekto tikslas - išspręsti robotikos uždavinį: tikslo paieškos navigacija naudojant aplinkos žymėjimą. Sukurta sistema imituoja graužiko erdvinės navigacijos procesą naudojant du skirtingus jutimo sensorius: dirbtinę regą ir dirbtinę uoslę. Projekte pristatoma sukurta sistema ir realizuoti erdvinės navigacijos algoritmai naudojantys vaizdinę ir kvapo informaciją. Aprašoma eksperimento vykdymo metodika ir priemonės. / The aim of the master thesis is to implement, rodent’s biological, goal directed navigation based on self-marking process in robotics. Rodents in environment navigate using analyzed information that gets from five different senses. Seeing helps to realize environment’s space, relief, distance from the objects. Hearing sense gives information about sound position of sound sources or other animals. Touch sense helps to determine properties of the surface: is it cold or hot, dry or wet, soft or hard. Olfactory and flavor receptors are used to determine food source or territory range. In real biological world animals could easily analyze and combine information from different sense receptors in spatial navigation process. In artificial system this data we are getting from different type and characteristics sensors. The main aim of this project is to solve robotic task: goal directed navigation based on self -marking method. Created system simulates rodent’s spatial navigation process where in two sensors are used: artificial eyes and artificial nose. This project presents created system and implemented spatial navigation algorithms which are using video and odor tracking based information. The methodology and tools used in experiment are described in this paper.

This work's aim is movement detection in the environment of a robot, that may move itself. A 2D occupancy grid representation is used, containing only the currently visible environment, without filtering in time. Motion detection is based on a grid-based particle filter introduced by Tanzmeister et al. in Grid-based Mapping and Tracking in Dynamic Environments using a Uniform Evidential Environment Representation. The system was implemented in the Robot Operating System, which allows for re-use of modules which the solution is composed of. The KITTI Visual Odometry dataset was chosen as a source~of LiDAR data for experiments, along with ground-truth pose information. Ground segmentation based on Loopy Belief Propagation was used to filter the point clouds. The implemeted motion detector is able to distiguish between static and dynamic vehicles in this dataset. Further tests in a simulated environment have shown some shortcomings in the detection of large continuous moving objects.

Hlavným cieľom práce bolo vytvorenie softvérového riešenia založeného na neurónových sieťach, pomocou ktorého bolo možné detegovať človeka a následne ho nasledovať. Tento výsledok bol dosiahnutý splnením jednotlivých bodov zadania tejto práce. V prvej časti práce je popísaný použitý hardvér, softvérové knižnice a rozhrania pre programovanie aplikácií (API), ako aj robotická platforma dodaná skupinou robotiky a umelej inteligencie ústavu automatizácie a meracej techniky Vysokého Učenia Technického v Brne, na ktorej bol výsledný robot postavený. Následne bola spracovaná rešerš viacerých typov neurónových sietí na detekciu osôb. Podrobne boli popísané štyri detektory. Niektoré z nich boli neskôr testované na klasickom počítači alebo na počítači NVIDIA Jetson Nano. V ďalšom kroku bolo vytvorené softvérové riešenie tvorené piatimi programmi, pomocou ktorého bolo dosiahnuté ciele ako rozpoznanie osoby pomocou neurónovej siete ped-100, určenie reálnej vzdialenosti vzhľadom k robotu pomocou monokulárnej kamery a riadenie roboty k úspešnému dosiahnutiu cieľa. Výstupom tejto práce je robotická platforma umožnujúca detekciu a nasledovanie osoby využiteľné v praxi.

Letecká fotogrammetrie v oblasti bezpilotních systémů představuje rychle rozvíjející se obor nalézající uplatnění napříč nejen průmyslovými odvětvími. Široce rozšířená metoda nepřímého georeferencování založená na vlícovacích bodech sice dosahuje vysoké přesnosti a spolehlivosti, v některých speciálních aplikacích nicméně není použitelná. Tato disertační práce se zabývá vývojem senzorického systému pro přímé georeferencování aplikovatelného na malých bezpilotních prostředcích a dále také návrhem vhodných kalibračních metod a testováním přesnosti. Významná část práce je věnována novým oblastem, kde může navržený systém pomoci eliminovat bezpečnostní rizika spojená s daným prostředím. V tomto kontextu byl systém testován v reálných podmínkách při mapování sněhu v horských oblastech a při robotickém mapování radiace.

The thesis deals with modelling and control of wheeled mobile robots. The thesis introduces the mathematical models of diferentially driven mobile robot, robot with Ackerman steering and a skid steered robot. The thesis describes open-loop and feedback control. Feedback control is divided into point-to-point control, path following and trajectory tracking.

This master thesis present a method for building topological maps for mobile robot navigation using neural network and neural fuzzy network. The master thesis concentrates on classification method. Neural fuzzy network is compared with two neural networks. It was also designed control algorithm exploration environment for autonomous mobile robot. This will rereduce the time to build the map. I developed simulation program in Matlab, which simulate move mobile robot in unknown environment.

Obsah Použité zkratky ...................................................................................................................... 9 Úvod .................................................................................................................................... 10 Výzkumný problém a cíle práce...................................................................................... 11 Využité metody práce...................................................................................................... 12 1 Teoretická východiska................................................................................................. 13 1.1 Informatické myšlení ............................................................................................ 14 1.2 Algoritmické myšlení ........................................................................................... 18 1.2.1 Algoritmus..................................................................................................... 19 1.2.2 Vlastnosti algoritmů ...................................................................................... 20 1.2.3 Způsoby zápisu algoritmů ............................................................................. 22 1.2.4 Základní algoritmické konstrukce...

This thesis deals with robot modelling and simulation of the robot behavior. At first we discuss a benefit of this kind of simulation. Next we describe the general scheme of robot. After that we select the environment where robot will be simulated in. There is a conception of this robot and explanation of the implementation in chosen environment. We continue with design and realization of the communication interface of this robot. The result of this procedure is robot with independent movement able to response to the obstructions and avoid them. Robot is able to create a map of the environment and use this map to find out the optimal way to the scheduled goal using the A* algorithm and achieve this goal. There is a possibility to use more robots in the scene. These communicate and the nearest one is able to get the goal.