|
|
Vectorized Point Clouds for Mobile Robotics
Jelínek, Aleš ; Mazal, Jan (referee) ; Duchoň,, František (referee) ; Žalud, Luděk (advisor)
Disertační práce se zabývá zpracováním mračenen bodů z laserových skenerů pomocí vektorizace a následnému vyhledávání korespondencí mezi takto získanými aproximacemi pro potřeby současné sebelokalizace a mapování v mobilní robotice. První nová metoda je určena pro segmentaci a filtraci surových dat a realizuje obě operace najednou v jednom algoritmu. Pro vektorizaci je představen optimalizovaný algoritmus založený na úplné metodě nejmenších čtverců, který je v současnosti patrně nejrychlejší ve své třídě a blíží se tak eliminačním metodám, které ovšem produkují výrazně horší aproxi- mace. Inovativní analytické metody jsou představeny i pro účely vyjádření podobnosti mezi dvěma vektorizovanými skeny, pro jejich optimální sesazení a pro vyhledávání korespondencí mezi nimi. Všechny představené algoritmy jsou intezivně testovány a jejich vlastnosti ověřeny množstvím experimentů.
|
|
| |
|
|
Measurement of laser source with a cylindrical lens
Jelínek, Aleš ; Šedivá, Soňa (referee) ; Bejček, Ludvík (advisor)
The purpose of this work is to design and create device for measuring the light intensity of laser source with cylindrical optics. It will be used to check laser source attributes during it's construction – primarily steady illumination along the whole line. The main goals were the minimization of construction complexity, reduction of cots and of course achieving all requirements. The product is fully functional machine, which works in company today.
|
|
|
Comparation of the 2D LIDAR SLAM methods in the simulation and in the real world
Cihlář, Miloš ; Jelínek, Aleš (referee) ; Ligocki, Adam (advisor)
This thesis deals with the design of a mobile robotic platform with the ability to build a map with an already existing 2D SLAM algorithm and a comparison of this algorithm in simulation and the real world. It uses three known algorithms i.e. Hector slam, Gmapping, and Karto slam. The part of this work deals with the creation of two wheels differential drive robot models in the Gazebo simulator and with an implementation proposal that is realized with the ROS framework. In thesis is implemented basic localization technique for mobile robots. Planning algorithms and a path controller are mentioned. In thesis is proposed own access to an autonomous exploration of building to build a map of the building. For evaluation of the SLAM algorithm is suggested slam quality criterion.
|
|
|
Linear Actuator with Stepper Motor
Farba, Marek ; Jelínek, Aleš (referee) ; Kopečný, Lukáš (advisor)
This bachelor's thesis deals with linear unit with stepper motor. The paper discusses design of the unit, a brief description of linear motion guides and the stepper motor. It also deals with analysis of the model proposed units, the calculation of torque stepper motor and simulation. The conclusion of the bachelor's thesis describes the selected components of the constructed linear unit and its process of manufacturing.
|
|
|
Windows 10 IoT on Raspberry Pi 2 computer
Prachař, Vojtěch ; Jílek, Tomáš (referee) ; Jelínek, Aleš (advisor)
The purpose of this thesis is to learn about the Raspberry Pi 2, Windows 10 IoT and development tools for this platform. In this work is presented hardware and software, and stated in detail the launching of development tools. In next part some sample applications and communication protocols are tested. In last part tested important ports for robotics. At the end, this platform is evaluated for the possibility of usage in home automation and robotics.
|
|
|
Leg Bone Posture Analysis Using Microsoft Kinect One
Tatýrek, Ondřej ; Jelínek, Aleš (referee) ; Chromý, Adam (advisor)
This master thesis is engaged in development of software tool intended for measuring of status of lower limb bones. It analyzes status of the feet and lower limb and defines significantpointsonthem.Alsoanalyzesdesignoptionsofmeasuringsceneandmarkers and describe the one selected. In another part is described automated point recognition algorithm. The last part describes graphical user interface, where are these algorithms implemented.
|
|
|
Unit for wireless transmission of electrical energy
Jelínek, Aleš ; Kopečný, Lukáš (referee) ; Žalud, Luděk (advisor)
The aim of this work is to prove the practical applicability of the concept of contactless energy transfer using induction coils. The first part deals with the mathematical model and theoretical proof of the concept being able to work in practice and describes the process of detecting resonance of receiver circuit on transmitter side. Next chapter describes hardware solution of the experimental prototype. The last part contains measurement results of a coupling coefficient of coils and discusses performance and notable features of the prototype.
|
|
|
Algorithms for browsing 2D maze
Chvíla, Lukáš ; Jelínek, Aleš (referee) ; Žalud, Luděk (advisor)
This thesis describes an automatic path finding method in a 2D maze defined by bitmap evidence grids. The problem was solved in the programming language C# with the help of the straight line decomposition algorithm used for creating a network of potential routes. The algorithm assumes approximation of obstacles by vector polygons. The resulting path was found by the means of algorithm A*. The result is a path described by line segments. After some modifications, the path could be used for robot navigation.
|
|
|
Transfer Learning For Deep Convolutional Neural Network From Rgb To Ir Domain
Ligocki, Adam ; Jelínek, Aleš
In this paper, we are presenting a proof of concept of our system for training of the YOLOv3 neural network for object detection of vehicles in thermal camera images. Our approach is unique in the way we are using a dataset containing a large number of synchronized range measurements as well as RGB and thermal images. We are using the existing YOLO toolkit to detect objects on the RGB images, we estimate detection distance by the LiDAR and later we reproject these detections into the IR image. In this way, we have created a large dataset of annotated thermal images that helped us to significantly improve the performance of the neural network at the IR domain.
|
guest ::
RSS feed.