|
|
Navigation of UAV in GNSS denied area
Pintér, Marco ; Marcoň, Petr
This paper examines the concept of navigation of of Unmanned aerial vehicle (UAV) in three-dimensional space using visual odometry. In the near future navigation of the UAV without GNSS is becoming a critical part of autonomous navigation systems, using information from on-board cameras to estimate the UAV’s movement and position. In the paper, different types of visual odometry, sensors for visual odometry, components of the implementation, and scenarios of usage. For the development and future application we utilize widely used Robotic Operating System (ROS).
|
|
|
The Human-machine interface for UAV ground control station
Klouda, Jan ; Marcoň, Petr
This paper describes the design of a Humanmachine interface for UAVs and presents general HMI requirements for the operator (commander). The design was created for a ground control station that is tasked with safely and efficiently controlling several UAVs at once. The goal was to create an environment that provides the operator with an overview of each UAV, mission planning and execution.
|
|
|
Obstacle Avoidance in UAVs: Using a Bug-Inspired Algorithm and Neural Network-Based RGB Camera Collision Prediction
Raichl, Petr ; Marcoň, Petr ; Janoušek, Jiří
Unmanned Aerial Vehicles (UAVs) are increasingly deployed in complex environments for various applications, necessitating advanced obstacle avoidance capabilities to ensure safety and mission success. Inspired by the simplicity and effectiveness of biological navigation strategies, this study introduces a novel approach to UAV obstacle avoidance, leveraging the principles of the bug algorithm combined with the predictive power of neural networks. We propose a hybrid model that integrates a lightweight neural network to predict potential collisions in real-time. Our methodology employs a two-stage process: first, the neural network assesses the immediate risk of collision; second, the bug algorithm-inspired decision-making process determines the UAV’s maneuvering actions to navigate without crashing to obstacles. The approach was tested both in simulation and real outdoor experiments.
|
|
|
Image data transmission system from UAV
Plaček, Jan ; Marcoň, Petr (referee) ; Janoušek, Jiří (advisor)
This bachelor thesis explores the processing of image data in the robotic operating system ROS2 for wireless transmission from unmanned aerial vehicles over mobile networks. It focuses on optimizing image data for fast transmission to cloud storage, emphasizing latency minimization and optimal frame rate depending on the environment type. The results of the thesis contribute to the real-time utilization of unmanned aerial vehicles, particularly in the field of robotic vision.
|
|
|
Visual-inertial odometry for UAV stabilization
Pintér, Marco ; Janoušek, Jiří (referee) ; Marcoň, Petr (advisor)
This master's thesis deals with the development of a system for estimating the position of an unmanned aerial vehicle in space without using GNSS (Global navigation satellite system). Visual and inertial data are employed for the position estimation of the unmanned aerial vehicle in space. The thesis describes various methods of determining the position using visual and inertial odometry. In the subsequent part of the thesis, the selection of a suitable camera sensor and onboard computer is discussed. The choice of an appropriate operating system, ROS (Robot Operating System) distribution, and a suitable algorithm for processing visual and inertial data is also described. To acquire telemetry data and basic control of an unmanned vehicle in space, a ROS node was implemented, which was tested in the Gazebo simulation environment with PX4-SITL. With the designed system experiments of comparing the accuracy of position determination against GNSS was done in real enviroment. From the achieved results, it can be concluded that visual odometry achieves very good results compared to GNSS.
|
|
|
Asset Administration Shell implementation
Maslowski, Jakub ; Marcoň, Petr (referee) ; Benešl, Tomáš (advisor)
This master's thesis builds upon the author's bachelor's thesis and gradually summarizes insights regarding Industry 4.0, Digital Twins, and primarily AAS. It focuses on describing AAS with regard to the necessary compromises from the theoretical concept of AAS to its practical implementation in the manufacturing process. The extensive practical part deals with the implementation of AAS in decentralized production into Siemens S7-1500 series PLC systems with OPC UA servers, including an individual solution for prioritized queue application. The counterpart representing products is a central AAS Hub Client implemented in Python. The results are presented in the form of testing on a network of 29 virtual PLCs.
|
|
|
REMOTE SENSING-BASED ANALYSIS OF MULTISPECTRAL INDICES
Janoušek, Jiří ; Kačmařík,, Michal (referee) ; Nedělník,, Jan (referee) ; Marcoň, Petr (advisor)
The thesis discusses various perspectives on the analysis of multispectral images, emphasizing in particular the technique’s potential in remote sensing of the Earth. Within multispectral imaging, the author has explored diverse methods for scanning plants to examine the impact exerted by the invididual electromagnetic spectrum components on the reflectivity of the investigated vegetation. Regarding the relevant vegetation indices, one of the main activities lay in comparing the approaches employed to compute the indices from applicable multispectral images. To deliver general data on the state of the plants being grown, the research involved designing and structuring relevant indicators that enabled the author to correlate the multispectral indices. In this context, the final sections of the thesis present the spectral maps acquired from the multispectral camera images and expose the analyses centered on the resulting index values, the aim being to help determine reliably the optimum harvest time in agricultural crops. The output of the project outlined within the thesis is then embodied in novel linear regression models that facilitate defining the optimum such a time. Importantly, by extension, the author also proposes a compact methodology that covers the individual phases and steps between the vegetation imaging the crops and finding the optimum harvest time.
|
|
|
Data Mining Methods for Text Analysis
Kozák, Ondřej ; Marcoň, Petr (referee) ; Dohnal, Přemysl (advisor)
This bachelor thesis explores the current methodology and possibilities of text mining and the subsequent application of some methods. The thesis described methods for preprocessing, methods for converting text to vector space and methods for text analysis and discusses their possible applications. The different preprocessing methods were applied to the text and then the conversion to vector space was demonstrated using simple methods such as BOW, Bag of n-grams, TF-IDF or with machine learning methods which are FastText and GloVe. LSA, LDA, TextRank and cosine similarity methods were applied to the extracted vectors to extract information from the text.
|
|
|
Drone swarm control
Kolísek, Josef ; Marcoň, Petr (referee) ; Janoušek, Jiří (advisor)
This thesis focuses on the drone swarm control and the using in practice. It is about the interconnection of several control units of drones and subsequent coordination in the air. The work is divided into theoretical part and practical part. The practical part describes the possibilities of drone swarm flying and coordination of multiple drones. Possible ways of communication and transmission of telemetry data for control using the MAVLink protocol. The next part describes the possibilities of controlling multiple units from one central point and creating flight instructions. The practical part deals with the realization of the software for synchronized flight of two unmanned aircraft and their connection with the ground control station for the transmission of flight instructions. The last part of the work deals with evaluation of the control, compliance of the established flight plan and the proposal of other possibilities for improvement with the solution of deficiencies.
|
|
| |
guest ::
