PhD researcher · UAV robotics · Visual-Inertial Odometry
About Me
Hi! I’m Šimon Prokop, a PhD researcher at Brno University of Technology, currently working in the areas of autonomous robotics, UAVs, computer vision, and Visual-Inertial Odometry.
My research focuses mainly on reliable state estimation and autonomous navigation for small aerial robots operating in GNSS-denied environments.
I am currently interested in combining classical geometric methods such as Visual-Inertial Odometry / SLAM with adaptive and continual-learning approaches.
🔬 Research Interests
- Visual-Inertial Odometry
- Visual Odometry
- Autonomous UAV Navigation
- Computer Vision
- State Estimation
- ROS / ROS 2
- Embedded Robotics
🚀 Current Research
Adaptive Visual-Inertial Odometry for UAVs
My current research investigates robust Visual-Inertial Odometry for small multirotor UAVs operating indoors and in GNSS-denied environments.
The goal is to develop and evaluate a real-time VIO pipeline capable of handling:
- fast UAV motion,
- rapid rotations,
- low-texture environments,
- motion blur,
- changing illumination,
- dynamic objects,
- limited onboard computational resources.
The system is primarily built around OpenVINS, ROS 2, synchronized camera + IMU measurements, and onboard embedded computing.
Technologies:
OpenVINS ROS 2 C++ Python OpenCV Eigen Jetson
GitHub · Project Page · Results
UAV Tracking of Dynamic Objects
Another research direction explores autonomous UAV tracking of moving and deformable objects, such as tree branches affected by wind.
The project combines:
- VIO-based drone state estimation,
- visual object tracking,
- relative target pose estimation,
- adaptive control,
- continual learning.
The objective is to enable the drone to continuously adapt its motion while following a dynamically moving target.
Technologies:
OpenVINS Computer Vision Continual Learning UAV Control
VIO Evaluation & Benchmarking Toolkit
I am also developing tools for evaluating Visual-Inertial Odometry algorithms using datasets such as EuRoC MAV.
The evaluation pipeline includes:
- Absolute Trajectory Error (ATE),
- Relative Pose Error (RPE),
- trajectory alignment,
- tracking success rate,
- runtime measurements,
- visualization,
- automatic experiment logging.
The goal is to create a simple and reproducible framework for comparing multiple VIO configurations and algorithms.
Technologies:
Python ROS 2 OpenVINS EuRoC NumPy OpenCV
📰 News
2026-09
Visiting research stay at the University of Bristol.
2026-07
Started research on adaptive UAV tracking using Visual-Inertial Odometry and continual learning.
2026-06
Started development of a UAV VIO dataset and evaluation pipeline.
2026-06
Development of real-time Visual-Inertial Odometry experiments for embedded UAV platforms.
📚 Publications
Work in progress.
🎓 Education & Research
Brno University of Technology
PhD Researcher
Research areas:
- robotics,
- autonomous systems,
- computer vision,
- aerial robotics,
- Visual-Inertial Odometry.
University of Bristol
Visiting PhD Researcher — 2026
Research collaboration focused on autonomous UAV navigation, Visual-Inertial Odometry and adaptive robotic systems.
💻 Technical Skills
Programming
C++ Python Bash MATLAB
Robotics
ROS ROS 2 MAVROS ArduPilot
Computer Vision
OpenCV OpenVINS ORB-SLAM VINS
Embedded
NVIDIA Jetson Raspberry Pi Flight Controllers
Tools
Git Docker CMake Linux
📈 Research Goals
My long-term research goal is to develop autonomous robotic systems that can operate reliably in complex environments without depending on external positioning infrastructure.
I am particularly interested in combining:
geometric robotics + state estimation
📬 Contact
Feel free to contact me regarding research collaboration, robotics projects, Visual-Inertial Odometry or UAV research.
Email: 221488@vut.cz
GitHub: Github
LinkedIn: LinkedIn
Last updated: September 2026
See my projects or get in touch by email.