|
|
Vehicle Control via Reinforcement Learning
Maslowski, Petr ; Uhlíř, Václav (referee) ; Šůstek, Martin (advisor)
The goal of this thesis is a creation of an autonomous agent that can control a vehicle. The agent utilizes reinforcement learning that uses neural networks. The agent interprets images from the front vehicle camera and selects appropriate actions to control the vehicle. I designed and created reward functions and then experimented with hyperparameters setup. Trained agent simulate driving on the road. The result of this thesis shows a possible approach to control an autonomous vehicle agent using machine learning method in CARLA simulator.
|
|
|
Coalition Formation in Multiagent Systems Using Genetic Algorithms
Kučera, Tomáš ; Uhlíř, Václav (referee) ; Zbořil, František (advisor)
This thesis discusses the basics of software agents and the way they form the multiagent coalitions. Genetic algorithms are introduced as one of the methods of solving the coalition formation problem. MAPC 2018 competition is introduced, which inspired the final design and implementation of the solution by using the tools described. A demo project was created, in which agents communicate with the MASSim server and gather data which is then used as an input into the genetic algorithm. Its purpose is to assign the agents to the tasks based on the input data, so that the tasks can be accomplished in the most effective manner possible. The results of this algorithm are evaluated in experiments which are focused on the quality of the solutions found as well as the time required for the calculation.
|
|
|
Path Planning using Voronoi Graphs
Živčák, Adam ; Uhlíř, Václav (referee) ; Rozman, Jaroslav (advisor)
The main purpose of this bachelor thesis is to design and implement plugin for robot operating system, which will be used for path planning using Voronoi graphs. Path planning is executed in well known world, about which robot knows where obstacles are placed. Thesis contains overview of main concepts of robot operating system, description of Voronoi graphs and algorithms to construct them. In conclusion is placed comparison of implemented plugin for path planning with ROS integrated planners.
|
|
|
Path Planning for Robot Formations
Hornáček, Zdenko ; Uhlíř, Václav (referee) ; Rozman, Jaroslav (advisor)
Robotics has been experiencing great development in recent years. This development is due to constant technical progress and the effort to develop increasingly robust robots and to further develop co-operating groups of robots. These co-operating groups (robot formations) have many areas of use ranging from environmental mapping to search and rescue missions. This bachelor thesis deals with the design of the path planning algorithm for robot formations. It uses the already existing planning algorithm for single robot - the Dijkstra algorithm. Based on the path computed by this algorithm, the proposed algorithm will calculate the path for each robot so as to maintain the shape of the formation during its move to the goal.
|
|
|
Film Suggestions Based on CSFD User Profiles
Janko, Pavel ; Šůstek, Martin (referee) ; Uhlíř, Václav (advisor)
This thesis covers the topic of utilizing neural nets for recommending movies. The principle of using neural nets with machine learning and both the general and the advanced techniques of creating a recommender system are also covered in the thesis. The core of the thesis is the design, implementation and finally the evaluation of a system for movie recommendations based upon the data mined from the user profiles from the ČSFD (Czech-Slovak film database). In order to accomplish this goal the system utilizies an explicit factorization model based on collaborative filtering between items to predict an accurate rating that the user would presumably give to a movie after watching it. This thesis also describes the relation between dataset size and prediction accuracy and demonstrates this accuracy by analyzing user feedback.
|
|
|
Coalition Games in a Dynamic Multiagent Environment
Hamran, Peter ; Uhlíř, Václav (referee) ; Zbořil, František (advisor)
The essence of this bachelor thesis was to implement an algorithm for suitable coalition forming in multi-agent system. Suitable coalition denotes a coalition which is not formed in super-additive environment and so it depends on its size and members. Aim of this thesis was to adjust an algorithm for coalition bounds forming between agents in a way, that it is applicable in a multi-agent competition MASSIM. Key aspects are amount of messages sent between agents for coalition to form and also a time it takes to calculate all coalitional values required for such a process. Final scenario is where two competitive groups of agents compete for resources in a multi-agent system simulation.
|
|
| |
|
|
Design of Quadrocopter
Uhlíř, Václav ; Rozman, Jaroslav (referee) ; Hájek, Josef (advisor)
This student paper discusses basic concept of quadrocopter for purpose of academic platform for testing and development. Paper includes basic overview of construction, stabilization and controlling of quadrocopter. As result of this paper is presented concept of quadrocopter and its constructed prototype with included software console for wireless control under Android system.
|
|
|
Robot Positioning Based on Sensor Measurements
Čakloš, Ondrej ; Uhlíř, Václav (referee) ; Rozman, Jaroslav (advisor)
The goal of this thesis is to create a program, that will be receiving measurements from robot's sensors, provide sensor fusion and estimate position of robot based on this fusion. For solving I used knowledge about probabilistic robotics, robotic operating system, information fusion, filtering especially extended Kalman filter and robot localization. I created an application of extended Kalman filter as a result. Filter listen to messages from robot sensors, providing a sensor fusion and estimating position of the robot in environment. Filter can receive measurements from multiple sources. The estimated states have proven themselves reasonably accurate for successful robot localization in space.
|
|
|
User Interface for RoboDelivery Unit
Vostřejž, Tomáš ; Zbořil, František (referee) ; Uhlíř, Václav (advisor)
The aim of this thesis is to design and implement user interface for RoboDelivery service. In addressing the problem, attention is drawn to the fact that interaction with the delivery unit must be intuitive and cover the widest possible range of users, including users with visual, auditory or tactile disabilities. Furthermore, the designed interface is tested with the help of users. The result is an ODROID application.
|
guest ::
RSS feed.