Robotics and Computer Engineering - Master's theses

Selle kollektsiooni püsiv URIhttps://hdl.handle.net/10062/42116

Sirvi

Nüüd näidatakse 1 - 20 131

listelement.badge.access-status Avatud juurdepääs ,
Design and Characterization of an Ultra-Wideband Radio for Extended-Range Transmission
(Tartu Ülikool, 2025) Vaht, Oliver; Kalde, Jaanus, juhendaja; Tartu Ülikool. Loodus- ja täppisteaduste valdkond; Tartu Ülikool. Tehnoloogiainstituut
This thesis presents the design and characterization of a custom ultra-wideband radio system for extended-range communication. Existing commercial modules were found inadequate due to limited configurability and restricted frequency range. To overcome these issues, a UWB platform was developed, including a transceiver based on the DW1000 chip, a power amplifier using the Renesas F1485, and a pentagon-shaped antenna optimized for 3–5 GHz. All components were validated through measurements. Field tests were conducted in real outdoor conditions with a special frequency license. Maximum communication distance achieved was 2.63 km using UWB Channel 4. Bit error rate remained low, with packet loss becoming the main limiting factor at longer ranges. The results show that with custom hardware and correct configuration, UWB can be used effectively for extended-range transmission in real-world conditions.
listelement.badge.access-status Avatud juurdepääs ,
Concept of a Robotic Beehive Based on a Case Study with a Beekeeping Company
(Tartu Ülikool, 2025) Väljaots, Mihkel; Tartu Ülikool. Loodus- ja täppisteaduste valdkond; Tartu Ülikool. Tehnoloogiainstituut
Throughout history, beekeepers have developed new types of beehives to improve beekeeping efficiency and enhance honeybees' quality of life. In the past five years, the trend in innovation has shifted toward full automation, with many examples emerging. This thesis follows that direction by investigating one professional beekeeping company, using a case study as the primary research methodology. This thesis explores the research questions: 'Which parameters are inspected by beekeepers?' and 'How can the beehive be automated based on these parameters?'. Participant observation, the researcher's diary, a semi-structured interview and digital imaging were used as data collection methods to gather and validate information relevant to the automation of the beehive. A custom imaging device was developed and employed to acquire approximately 3000 frame images across more than 100 Langstroth beehives. These images served as input for evaluating Varroa destructor detection performance using the YOLO-based computer vision system. It was observed that the case study partner (whose core business is honey production) does not measure any specific parameters in the beehive to run a successful beekeeping business. Decisions and actions are based on the evaluation of 20 parameters, the most important of which are the queen's health, temperature, and available space for brood and honey. The greatest challenge in automating the beehive is the localisation of the queen. A magnetic tag and smart frames with digital magnetic sensors are proposed to address this issue. The second challenge involves moving frames that are stuck together, as breaking them loose can sometimes require more than 300 N of force. The third challenge is to move frames between beehive bodies while keeping honeybee losses as low as possible. The collected information was analysed, and the requirements for a robotic beehive are presented along with the overall concept of the robotic beehive. The case study partner describes the perfect beekeeper as someone who 'sees everything and remembers everything.' Since the partner relies on very general notes when making decisions in beehive management, and it is not feasible for humans to memorise images of thousands of frames, a robotic beehive equipped with computer vision and hive management functions is relevant from both scientific and commercial perspectives. It is recommended that the first development phase of robotic beehives be developed to explore possible visual patterns and sound or temperature signatures that could transform beekeeping operations in unprecedented ways.
listelement.badge.access-status Avatud juurdepääs ,
Development of a 3D web visualization component for KuupKulgur Mission Control Interface
(Tartu Ülikool, 2025) Pani, Hans Pärtel; Tartu Ülikool. Loodus- ja täppisteaduste valdkond; Tartu Ülikool. Tehnoloogiainstituut
This thesis presents the development and design of a web based 3D visualization component prototype for the KuupKulgur Lunar rover. Several existing robotics visualization software options were analyzed and the ideas incorporated into making this component. The software was built using well-established web development libraries like Lit and Three.js. The performance of the component was tested using several datasets of pre-collected data and tested in simulated scenarios.
listelement.badge.access-status Avatud juurdepääs ,
Programming of a Robotic Gantry for Material Position Correction
(Tartu Ülikool, 2025) Palacios Paniagua, Borja; Tartu Ülikool. Loodus- ja täppisteaduste valdkond; Tartu Ülikool. Tehnoloogiainstituut
Robotic gantry systems are essential in automation, offering the precision and adaptability needed for advanced manufacturing. This thesis focuses on programming a robotic gantry system for correcting material positions, an operation critical for ensuring quality in industrial processes like assembly and packaging. The problem arises from the challenge of detecting and aligning materials in real-time, especially given variations in material, lighting, and positioning errors. These issues directly impact the efficiency and reliability of the manufacturing process, necessitating advanced solutions. The proposed work addresses these challenges by leveraging: • A robust machine vision system for detecting material edges and positions accurately. • Kinematic models to calculate corrections in position and orientation. • Calibration methodologies for aligning vision systems with gantry reference frames. The project integrates image processing algorithms and systematic calibration steps to enhance the detection accuracy and system reliability. The work is particularly significant as it aligns with industry trends toward higher precision and automation.
listelement.badge.access-status Avatud juurdepääs ,
Development and Validation of a Lunar Rover Digital Twin Using Unreal Engine 5
(Tartu Ülikool, 2025) Mihailovs, Audris; Tartu Ülikool. Loodus- ja täppisteaduste valdkond; Tartu Ülikool. Tehnoloogiainstituut
This thesis will provide a concise report on the modern technologies connected with the space industry focusing on robotics and its development using copies of physical systems in simulation software or digital twins. It will go on to describe the software tools used and potential candidates for multipurpose use in the future and their respective pros and cons. Second section will be about the creation of the Lunar rover’s digital twin in Unreal Engine 5. It will discuss the whole process of building, implementing, and calibrating the digital twin to be as accurate as possible to the physical rover. As well as compare simulated sensor data with the physical sensors to understand the differences and limitations of the software used. This work provides the basis for a precise digital twin on to whom future work can be built upon and more advanced systems implemented.
listelement.badge.access-status Avatud juurdepääs ,
Haptic-Based Teleoperation of Robots
(Tartu Ülikool, 2025) Mandiola Arrizabalaga, Asier; Tartu Ülikool. Loodus- ja täppisteaduste valdkond; Tartu Ülikool. Tehnoloogiainstituut
This thesis has been developed at Tekniker, a technological research center located in the Basque Country, as part of the Autopilot project, which is focused on designing a robotic system to assist in surgical applications. The objective is to enable safe and intuitive teleoperation through a combination of force feedback and precise motion control. The work presents the development and implementation of a bilateral robot teleoperation system with haptic feedback using a PLC-based external control architecture. The system integrates key components such as a robotic manipulator, a simulated haptic device (Omega.7), force sensing, and real-time communication. A non-deterministic communication protocol (ADS and UDP) is used to enable data exchange between the haptic device and the robot controller via a PLC. The first milestone was the successful teleoperation of the robot using the haptic device simulation as a master interface, enabling smooth and accurate control of the slave manipulator. Once this was achieved, the integration of force feedback was addressed to close the bilateral control loop. Experimental results confirmed that force data is correctly acquired, processed, and scaled to provide realistic haptic sensations to the operator. Communication latency was identified as a critical factor affecting system stability and was successfully mitigated through protocol optimization and task scheduling. The overall system proved to be robust, modular, and adaptable. The architecture is fully prepared for the integration of the physical Omega.7 device and is intended to be tested in realistic environments in future work, with potential applications in areas such as surgical simulation or remote robotic manipulation
listelement.badge.access-status Avatud juurdepääs ,
Development of modular Electrical Power System for KuupKulgur prototype
(Tartu Ülikool, 2025) Lindmaa, Laur Edvard; Tartu Ülikool. Loodus- ja täppisteaduste valdkond; Tartu Ülikool. Tehnoloogiainstituut
CubeRovers are a new class of lightweight, modular robotic platforms designed to lower the barrier to lunar surface exploration. To support such systems, a reliable and modular Electrical Power System (EPS) is essential. This thesis focuses on the development of an EPS for KuupKulgur, a CubeRover prototype created at Tartu Observatory. An advanced EPS-v2 was designed to meet a comprehensive set of power and communication requirements. While EPS-v2 serves as the foundation for future implementations, a simplified EPS-v1.1 was subsequently developed based on lessons learned during the design of v2. This version was tailored to enhance the reliability and performance of the current KuupKulgur rover prototype. The work includes the design, prototyping, and testing of both EPS boards. All system requirements were fulfilled, with the exception of hot-swappability, which was deprioritized due to mechanical and reliability constraints. Although space qualification was beyond the scope of this thesis, the system architecture allows for potential future thermal-vacuum and radiation testing. The results presented here lay the groundwork for future EPS revisions compatible with CubeRover standards and operational lunar missions.
listelement.badge.access-status Avatud juurdepääs ,
Video Compression Effect on a Camera-based Occupancy Prediction Model
(Tartu Ülikool, 2025) Leppsalu, Jürgen; Tartu Ülikool. Loodus- ja täppisteaduste valdkond; Tartu Ülikool. Tehnoloogiainstituut
Camera-based occupancy prediction models estimate the presence and distance of obstacles using camera images and camera positional data. This thesis investigates how lossy video compression impacts the performance of such a model. To test this, video compression was applied using three video codecs: MJPEG, H.264, and H.265. The compressed frames were then used to train 15 versions of an occupancy prediction model, and the resulting validation loss was analyzed for each configuration. The findings show that the selected occupancy prediction model demonstrated a high degree of resilience to compression artefacts. Although high-compression variants produced less stable predictions across consecutive frames, the predictions still contained mostly valid structures that resembled the expected outcome. As this study was conducted using synthetic data generated in the CARLA simulation environment, future research should explore whether models trained on real-world datasets exhibit similar robustness.
listelement.badge.access-status Avatud juurdepääs ,
Natural Language Human-Robot Interaction: A Modular Framework for Conversational Robot Control using Large Language Models
(Tartu Ülikool, 2025) Leclerc, Julian Rene; Tartu Ülikool. Loodus- ja täppisteaduste valdkond; Tartu Ülikool. Tehnoloogiainstituut
As robots are increasingly used in everyday situations beyond industrial environments, traditional controller-based interfaces that require specialized training pose significant barriers to widespread adoption. This thesis introduces a modular natural language human-robot interaction system designed for conversational robot control. It integrates large language models with robust robotic frameworks to enhance usability. The system tackles key challenges, such as the reliability of large language models, contextual awareness, and bridging the semantic gap. It includes specialized components for error handling, memory management, and cross-platform deployment through a web interface. Validated on various platforms, including TIAGo and Boston Dynamics Spot, the system demonstrates platform-agnostic functionality and accessibility for non-technical users. This advancement helps democratize robot control through natural language while ensuring operational reliability.
listelement.badge.access-status Avatud juurdepääs ,
Model-based planning using GPU-accelerated Simulator as a World Model
(Tartu Ülikool, 2025) Hurova, Iryna; Singh, Arun Kumar, juhendaja; Kruusamäe, Karl, juhendaja; Tartu Ülikool. Loodus- ja täppisteaduste valdkond; Tartu Ülikool. Tehnoloogiainstituut
Manipulator robots are increasingly deployed in real-world tasks that require smooth, reactive motion and robust collision avoidance, particularly in dynamic and unstructured environments. This thesis presents a model-based, collision-free, online trajectory optimization framework tailored for such scenarios. The method involves sampling hundreds of trajectories from a multivariate normal distribution, shaping them with Bernstein polynomials, and evaluating them in parallel within a MuJoCo simulation. These trajectories are then optimized using the cross-entropy method. The system achieves real-time, in-the-loop planning by integrating a model predictive control strategy. The experiment, both in simulation and in real-world tests, demonstrated successful manipulation in an environment with multiple obstacles. In addition, the framework supports flexible task objectives by adjusting the cost function, enabling goal-driven behavior under varying conditions.
listelement.badge.access-status Avatud juurdepääs ,
SurfMotion: An Open Source Pipeline for Robotic Pipe Cutting and Welding
(Tartu Ülikool, 2025) Hult, Carl Hjalmar Love; Tartu Ülikool. Loodus- ja täppisteaduste valdkond; Tartu Ülikool. Tehnoloogiainstituut
Robotic pipe cutting and welding are examples of precise surface-conforming tasks for manipulator robots. Though they are commonplace industrial processes, they pose difficult technical challenges, particularly in motion planning and execution on complex geometries. Existing solutions are largely proprietary, expensive, and closely tied to specific hardware platforms, while open-source alternatives are limited in scope and integration. This thesis introduces SurfMotion, a ROS package developed to begin to bridge the gap between open-source and proprietary solutions for surface-conforming robotic tasks. SurfMotion provides a modular pipeline for generating, projecting and executing geometry-aware Cartesian trajectories on complex surfaces. The package integrates surface processing, path projection, inverse kinematics feasbility checking, and robot-agnostic trajectory execution, leveraging the ROS 2 and MoveIt frameworks. The effectiveness of SurfMotion is demonstrated and quantified in terms of Cartesian space error, velocity consistency, and overall smoothness of motion execution through simulated robotic cutting and welding tasks across four example robot models (UR5, UR20, Fanuc m10ia and Kuka iiwa14). Experimental results show that two of the MoveIt motion execution frameworks integrated achieve sub-millimeter accuracy in path tracking.
listelement.badge.access-status Avatud juurdepääs ,
Designing Multimodal Emotional Expression for a Robotic Study Companion
(Tartu Ülikool, 2025) Calafa’, Miriam; Tartu Ülikool. Loodus- ja täppisteaduste valdkond; Tartu Ülikool. Tehnoloogiainstituut
This thesis investigates the design, implementation, and evaluation of the emotional expression of the Robot Study Companion (RSC) intended to support university students’ motivation, wellbeing, and learning outcomes. Guided by a theoretical review of affective feedback in educational robotics, we identified six target emotions: Anger, Joy, Pride, Fun, Surprise, and Caring, each linked to educationally meaningful functions. We developed a fully functional digital twin of the RSC in ROS 2 to enable rapid prototyping and simulation, and built a modular emotional expression that coordinates color, motion, facial expression, and vocal behavior for each emotional response. To assess recognition accuracy and user experience, we conducted a video-based cross-cultural study with 47 participants. Participants distinguished positive versus negative valence with over 95% accuracy, and achieved an overall emotion-recognition rate of 58.16%. User Experience Questionnaire (UEQ) results were consistently positive, with Perspicuity and Attractiveness rated highest. Qualitative interviews highlighted the strengths of our multimodal design as well as opportunities for refinement. Cross-group analyses revealed systematic differences by gender, academic level, and nationality, underscoring the need for adaptive, personalized emotion profiles. All software developed in this work is released open-source, providing a robust foundation for future research in affective human–robot interaction within educational settings.
listelement.badge.access-status Avatud juurdepääs ,
LTE-põhise passiivradari simuleerimine
(Tartu Ülikool, 2025) Avi, Mattias; Tartu Ülikool. Loodus- ja täppisteaduste valdkond; Tartu Ülikool. Tehnoloogiainstituut
Antud magistritöö käigus uuris autor passiivradarisüsteemide toimimise, olemasolevate süsteemide ja simuleerimise võimaluste kohta. Töö annab ülevaate passiivradari geomeetrilistest omapäradest ning muudest eelistest ja puudustest. Samuti veendutakse LTE tehnoloogia sobivuses passiivradari tööks. Tuuakse välja mõned märkimisväärsed eksperimendid, kus LTE tehnoloogial põhinev passiivradar on edukalt kontrollitud tingimustes suutnud tuvastada erinevaid objekte. Passiivradari süsteemi osasid uuriti, et selgitada välja milliseid on mõistlik simuleerida, et passiivradari arendamist lihtsustada. Seejärel töötati välja vahendid, mis suutsid seda ülesandeid täita ideaalsetes tingimustes.
listelement.badge.access-status Avatud juurdepääs ,
Socially Aware Planning for Indoor Navigation
(Tartu Ülikool, 2025) Avalos Conchas, Paola; Kruusamäe, Karl, juhendaja; Zhang, Bin, juhendaja; Tartu Ülikool. Loodus- ja täppisteaduste valdkond; Tartu Ülikool. Tehnoloogiainstituut
As robots increasingly present in human-populated spaces, they must be able to navigate among humans safely and without disrupting. People try to preserve their own personal space when moving in real-world social spaces. However, the current navigation methods do not consider this aspect and treat humans as any other obstacle.This thesis proposes a method that considers personal space for humans as well as social navigation norms. For this purpose, the robot converts camera-based human detections to a costmap form and define the personal space as a Gaussian asymmetric function. The proposed solution is validated through real-world experiments, demonstrating that the robot can improve the quality of navigation. The proposed solution is available on GitHub as a costmap layer that can be easily integrated into existing frameworks.
listelement.badge.access-status Avatud juurdepääs ,
Risk-Aware Planning on Point Clouds
(Tartu Ülikool, 2025) Akuamoah Boateng, Kwasi; Tartu Ülikool. Loodus- ja täppisteaduste valdkond; Tartu Ülikool. Tehnoloogiainstituut
Navigating complex environments safely is a critical challenge for autonomous drones. This thesis introduces a novel risk-aware planning framework that empowers drones to make smarter, safer decisions. At its core, the framework utilizes an innovative ensemble of neural networks (integrating PointNet, a point cloud processing network, and Gaussian policy-based Multi- Layer Perceptron (MLP) structures) to deliver probabilistic predictions of obstacle distances and their associated uncertainties. Coupled with a jerk-controlled trajectory model, the system leverages Conditional Value-at-Risk (CVaR) and a cross-entropy optimization method to intelligently quantify and mitigate risky trajectories. This allows the drone to confidently navigate by optimally balancing mission objectives against this principled risk measure. Simulation results demonstrate the effectiveness of the proposed framework.
listelement.badge.access-status Avatud juurdepääs ,
INTUIT-VLNCE: Autonomous Navigation through Vision-and-Language
(Tartu Ülikool, 2024) Rodionov, Kirill; Roy, Kallol; Tartu Ülikool. Loodus- ja täppisteaduste valdkond; Tartu Ülikool. Tehnoloogiainstituut
The aim of this thesis was to developed an Embodied Agent with explicit intuition capable of navigating indoor Continuous Environments based on provided Natural Language instruction and Agent’s egocentric vision as part of a Vision-and-Language Navigation task. The thesis proposes creating explicit intuition by making an Agent predict not only an action to perform at a given time, but also predicting actions for the future. An Agent’s policy was trained mimicking the training procedure from LAW-VLNCE project [1]. Evaluations showed negative results after implementing proposed method.
listelement.badge.access-status Avatud juurdepääs ,
VR-Enhanced Remote Inspection Framework for Semi-Autonomous Robot Fleet
(Tartu Ülikool, 2024) Reynes, Gautier; Valner, Robert; Norbisrath, Ulrich; Tartu Ülikool. Loodus- ja täppisteaduste valdkond; Tartu Ülikool. Tehnoloogiainstituut
This thesis presents the design and development of a Virtual Reality (VR)-enhanced user interface and communication infrastructure for remote inspection using a semiautonomous robot fleet. The core of this project is the creation of a VR interface that allows operators to immerse themselves in a digital twin of the remote environment, facilitating intuitive and efficient control over robot inspection. This interface supports both third-person and robot’s point-of-view perspectives, enhancing situational awareness and decision-making capabilities in hazardous environments. The software framework is built upon ROS 2 Foxy, and the VR application was designed with a new graphics engine called Wonderland Engine, particularly suited for lightweightWebXR experiences capable of running on a number of VR headsets, like the Oculus Quest 2. The communication between the interface and the robot fleet is tackled by a custom-built WebSocket server. The work is demonstrated using simulated robot scenarios in Gazebo. The demonstration serves as a proof of concept, showcasing the viability of the VR interface in a controlled environment and setting the stage for future real-world applications. This work contributes to the field of VR-enhanced remote inspection by providing an interface that bridges the gap between operators and remote environments. The integration of VR technology with robotic systems opens new possibilities for remote operation, offering a more immersive and intuitive control mechanism that can be adapted to various industrial and research applications.
listelement.badge.access-status Avatud juurdepääs ,
Continuous Collaborative Mapping in Unknown Environments: A Multi-Robot System Approach
(Tartu Ülikool, 2024) Radigon, Malcom; Muhammad, Naveed; Tartu Ülikool. Loodus- ja täppisteaduste valdkond; Tartu Ülikool. Tehnoloogiainstituut
Navigating and exploring unknown terrains remains a critical challenge within the field of mobile robotics. Achieving rapid and consistent exploration is crucial for the prompt generation of precise maps. This thesis introduces a cutting-edge distributed exploration system utilizing a multi-robot fleet. This innovative system is crafted to facilitate continuous exploration in unexplored areas by implementing a novel dronebased communication relay method. Additionally, it enables the synthesis of an integrated, comprehensive global map, providing crucial, rapid insights for human operators into the explored areas. The effectiveness of this system has been thoroughly evaluated through a series of simulated experiments encompassing various trials. These evaluations underscore the system’s capability in smoothly conducting exploration tasks, notably overcoming delays traditionally linked to communication challenges.
listelement.badge.access-status Avatud juurdepääs ,
Small-scale cars for autonomous driving research
(Tartu Ülikool, 2024) Petrović, Uroš; Muhammad, Naveed; Tartu Ülikool. Loodus- ja täppisteaduste valdkond; Tartu Ülikool. Tehnoloogiainstituut
Self-driving cars (autonomous cars, robo cars) are vehicles that can drive without human input. In the recent decades, this field of science has received a lot of attention. There is a growing need to improve the safety and effectiveness of traffic. Self-driving cars have the promise of being safe. The main reasoning is that by removing human error, and utilising algorithms, the number of mistakes and accidents that would arise in most situations could be drastically reduced. Autonomous driving is a heavily investigated area currently, but significant challenges remain open, thus requiring significant research during the decades to come. As the cost of life-size test platforms for autonomous-driving research is very high, the need to use smaller-scale vehicles to be employed as test platforms arises. This thesis investigates the viability of such a method, by evaluating the performance and suitability of a small-scale self-driving car platform, the Donkey Car, for different aspects of autonomous driving research. Perception, localization and mapping, planning and end-to-end capabilities of the platform were investigated and compared with publicly available real-life self-driving car experiments. The results achieved showed that small-scale cars are capable of performing autonomous driving research at an acceptable level in comparison with real-life cars.
listelement.badge.access-status Avatud juurdepääs ,
PHOENIX: Revisiting Cloudlet Development with Recycled Phones.
(Tartu Ülikool, 2024) Ngoy, Perseverance Munga; Flores, Huber; Norbisrath, Ulrich; Tartu Ülikool. Loodus- ja täppisteaduste valdkond; Tartu Ülikool. Tehnoloogiainstituut
This work revisits the idea of re-purposing e-waste (aka old electronics) into cloudlets that can be used as general computing units for several applications, including edge computing, the Internet of Things (IoT), and pervasive computing applications. While the idea has been around for over a decade, the continuous evolution of personal electronics makes it difficult to keep up to date with general guidelines and principles for re-purposing. As a result, we investigate the latest developments in this topic and provide newer insights for recycling deprecated electronics using modern tools and frameworks. To do this, four Nexus 5 phones are adapted with PostmarketOS, a Linux-based mobile OS, and subjected to benchmark tests alongside a Raspberry Pi 4 and a laptop. Results indicate lower performance metrics in mobile phones, accompanied by substantial heat generation during intensive tasks. The study delves into the advantages, disadvantages, and techniques for optimization, including load balancing, task scheduling, parallel computing, cloud offloading, and energy-aware algorithms. Potential applications are explored, emphasizing community empowerment, low-cost microclouds, and disaster response. While mobile phone-based cloudlets show promise, the challenges and limitations must be addressed. The study underscores the environmental benefits of repurposing mobile phones and proposes sustainable practices, including the removal of batteries for a reduced carbon footprint. The thesis also advocates for the educational empowerment of underprivileged students through accessible technologies, highlighting the potential societal impact of this innovative approach to edge computing.

Sirvi

Viimati lisatud