|
|
Road detection for mobile robot using image processing
Coufal, Jan ; Štarha, Pavel (referee) ; Krejsa, Jiří (advisor)
Diploma thesis deals with image processing for outdoor environment mobile robot. In first part, the problem is analyzed, general solution is proposed and suitable image processing methods are presented. In second part presented methods are tested and methods with best results are proposed. In third part is particular solution tested on real data.
|
|
|
Mobile Robot Localization in Indoor Scenarios
Sýkora, Tomáš ; Brejcha, Jan (referee) ; Beran, Vítězslav (advisor)
The goal of this project was to create a software for an autonomous mobile robot capable of navigation in indoor scenarios and of detection and recognition of the objects on its way. An important condition that the final software had to meet was to find the cheapest solution possible. I achieved the given goal and its conditions using an open source software framework Robotic Operating System (ROS) and its tools. The final system is a set of ROS modules communicating with each other, using a single sensor as an input data stream and a small motor which is able to send information about its velocity and rotations to the control part of the system. This mobile robot can, without any significant problems, navigate through a mapped area while trying to find trained objects. Such a robot could be easily used (after adding specific mechanical parts) to help retired or disabled people, to cooperate with industry workers or in many other fields.
|
|
|
Interest-Point Detection on CUDA
Ryba, Jan ; Řezníček, Ivo (referee) ; Herout, Adam (advisor)
Corner point detection is one of many functions used in computer vision for tasks such as tracking, detecting objects, comparing images and much more. Many of the algorithms are complex and require a lot of CPU time. This is where the CUDA platform comes in. CUDA kernels run parallely on graphic accelerators can rapidly decrease time needed for execution, allowing even these complex calculations to work in real time or even better. Text focuces on Moravec and Harris corner detection algorithms and their effective implementation on CUDA. Examination of potetntial and performance of CUDA platform is also importatnt.
|
|
|
Searching for Points of Interest in Raster Image
Kaněčka, Petr ; Sumec, Stanislav (referee) ; Herout, Adam (advisor)
This document deals with an image points of interest detection possibilities, especially corner detectors. Many applications which are interested in computer vision needs these points as their necessary step in the image processing. It describes the reasons why it is so useful to find these points and shows some basic methods to find them. There are compared features of these methods at the end.
|
|
|
Mobile platform path planning algorithm design
Smolinský, Michal ; Kůdela, Jakub (referee) ; Holoubek, Tomáš (advisor)
The aim of this master's thesis is to create a program of selected laboratory tasks for the mobile robot TurtleBot3 Burger. These tasks are the mapping of an unknown environment and the design of a path planning algorithm using the ROS framework. The first chapters of the work, which describe the field of mobile robotics, the TurtleBot3 Burger robot, the Robotic Operating System, the used path planning algorithms, and The ROS Navigation Stack, are devoted to clarifying the necessary knowledge. After the robot is put into operation, the mapping and path planning task is created. The following chapter is dedicated to verifying the functionality of both tasks, which also includes a comparison of the used path planning algorithms. The evaluation of the application of the algorithms on the TurtleBot3 Burger platform, the benefits of the proposed algorithms and their practical use in a real environment is described in the final chapter of this work.
|
|
|
Mobile Robot Localization in Indoor Scenarios
Sýkora, Tomáš ; Brejcha, Jan (referee) ; Beran, Vítězslav (advisor)
The goal of this project was to create a software for an autonomous mobile robot capable of navigation in indoor scenarios and of detection and recognition of the objects on its way. An important condition that the final software had to meet was to find the cheapest solution possible. I achieved the given goal and its conditions using an open source software framework Robotic Operating System (ROS) and its tools. The final system is a set of ROS modules communicating with each other, using a single sensor as an input data stream and a small motor which is able to send information about its velocity and rotations to the control part of the system. This mobile robot can, without any significant problems, navigate through a mapped area while trying to find trained objects. Such a robot could be easily used (after adding specific mechanical parts) to help retired or disabled people, to cooperate with industry workers or in many other fields.
|
|
|
Interest-Point Detection on CUDA
Ryba, Jan ; Řezníček, Ivo (referee) ; Herout, Adam (advisor)
Corner point detection is one of many functions used in computer vision for tasks such as tracking, detecting objects, comparing images and much more. Many of the algorithms are complex and require a lot of CPU time. This is where the CUDA platform comes in. CUDA kernels run parallely on graphic accelerators can rapidly decrease time needed for execution, allowing even these complex calculations to work in real time or even better. Text focuces on Moravec and Harris corner detection algorithms and their effective implementation on CUDA. Examination of potetntial and performance of CUDA platform is also importatnt.
|
|
|
Searching for Points of Interest in Raster Image
Kaněčka, Petr ; Sumec, Stanislav (referee) ; Herout, Adam (advisor)
This document deals with an image points of interest detection possibilities, especially corner detectors. Many applications which are interested in computer vision needs these points as their necessary step in the image processing. It describes the reasons why it is so useful to find these points and shows some basic methods to find them. There are compared features of these methods at the end.
|
|
|
Road detection for mobile robot using image processing
Coufal, Jan ; Štarha, Pavel (referee) ; Krejsa, Jiří (advisor)
Diploma thesis deals with image processing for outdoor environment mobile robot. In first part, the problem is analyzed, general solution is proposed and suitable image processing methods are presented. In second part presented methods are tested and methods with best results are proposed. In third part is particular solution tested on real data.
|
guest ::
