MSC SYSTEMS AND CONTROL — TU DELFT

PanagiotisIoakeimidis

Building autonomous robotic systems with ROS, state estimation, and real-time perception — from EKF-based navigation stacks to embedded TinyML control systems.

Delft, Netherlands
RÉSUMÉ

Selected Works

EKF Robot Localization
View on GitHub

EKF Robot Localization

ROS · C++ · SENSOR FUSION

01

ROS Noetic two-stage localization stack combining a map/GPS-aware global EKF with a high-rate odom/IMU local EKF. Includes Gazebo launch files, URDF robot models, and tuned robot_localization configs for simulation and real-robot deployment.

TinyML Self-Balancing Robot
View on GitHub

TinyML Self-Balancing Robot

C++ · TENSORFLOW LITE · LQR

02

TinyML implementation of a self-balancing robot on the Arduino Nano 33 BLE Sense using TensorFlow Lite Micro. Training data generated from an LQR-based Full-State Feedback controller in MATLAB/Simulink.

Lane Following Simulation
View on GitHub

Lane Following Simulation

PYTHON · OPENCV · COMPUTER VISION

03

Python and OpenCV-based lane following system that detects road lane markings from video, estimates the steering angle in real time, and outputs a normalized steering command using classical computer vision techniques.

Sign & Traffic Object Detection
View on GitHub

Sign & Traffic Object Detection

PYTHON · YOLO · OPENVINO

04

Real-time traffic sign, traffic light, pedestrian, and vehicle detection using YOLOv8 and OpenVINO. Processes video input with optimized inference, overlays detections with FPS metrics, and generates autonomous driving action cues.

Autonomous Navigation Stack
IN PROGRESSView on GitHub

Autonomous Navigation Stack

ROS · C++ · MOTION PLANNING

05

ROS1 Noetic autonomous navigation system for a differential drive robot, combining EKF localization with the ROS Navigation Stack. Features global (NavFn) and local (DWA) path planning in Gazebo with occupancy grid maps and depth camera-based obstacle perception.

ABOUT ME

Engineering autonomous systems at the intersection of perception and control.

I'm Panagiotis, a robotics software engineer focused on autonomous systems that perceive, localize, and navigate in real-world environments.

I work at the intersection of sensor fusion, computer vision, and motion control, building systems that connect perception to decision-making and execution.

My experience includes ROS-based navigation stacks, EKF sensor fusion, Visual SLAM, and deep learning-based perception pipelines deployed on embedded hardware using OpenVINO and TensorFlow Lite Micro.

I enjoy building systems that are not just algorithmically correct, but stable, modular, and deployable in real robotic environments.

Panagiotis Ioakeimidis

PANAGIOTIS IOAKEIMIDIS

Technical Arsenal

PROGRAMMING & SYSTEMS

  • C++
  • Python
  • MATLAB/Simulink
  • Bash
  • Linux
  • SQL

ROBOTICS & MIDDLEWARE

  • ROS1
  • Gazebo
  • RViz
  • MapViz
  • TF2
  • URDF

AUTONOMY & PERCEPTION

  • SLAM
  • State Estimation
  • EKF
  • YOLOv8
  • OpenCV
  • Point Clouds
  • Depth Cameras
  • Visual SLAM

MOTION PLANNING & CONTROL

  • LQR
  • MPC
  • NavFn
  • DWA
  • Lane Following
  • Full-State Feedback

EMBEDDED & EDGE ML

  • TensorFlow Lite Micro
  • OpenVINO
  • Arduino
  • TinyML
  • Raspberry Pi

TOOLS & WORKFLOW

  • Git
  • GitHub
  • GitLab
  • Docker
  • CMake
  • Make
EXPERIENCE

Robotics Engineer @ Beyond Orbit Team

NATIONAL TECHNICAL UNIVERSITY OF ATHENS

  • Developed a hybrid localization system combining Visual SLAM with GPS, IMU, and wheel odometry using EKF-based sensor fusion for robust state estimation in GPS-degraded environments.
  • Engineered autonomous navigation stack using ROS (move_base), integrating NavFn global planning and DWA local planning for reliable waypoint execution.
  • Designed depth-camera-based traversability and elevation mapping pipeline for terrain understanding and obstacle-aware navigation.
  • Implemented fault-tolerant localization architecture with fallback fusion of GPS, IMU, and odometry to ensure continuity during SLAM degradation.
  • Built RViz-based mission control interface with satellite map overlays enabling real-time localization, path monitoring, and operator situational awareness.

December 2026 — Present

Robotics Software Engineer Intern @ QSC

ATHENS

  • Developed a ROS1-based AR simulation environment in Blender for testing robotics perception and human-robot interaction in a controlled virtual setting.
  • Designed and implemented a multimodal sensor fusion pipeline combining vision and spatial audio for robust environment perception.
  • Integrated ceiling microphone arrays for 3D sound source localization, enhancing situational awareness in simulated robotics scenarios.

February 2026 — August 2026

Autonomous Systems / Computer Vision Engineer @ Bosch Future Mobility Challenge

BOSCH

  • Developed a robust lane-following algorithm using the Hough Transform, improving vehicle path estimation and navigation accuracy.
  • Designed, trained, and fine-tuned a YOLOv8-based traffic light and traffic sign detection system using Ultralytics, including dataset preparation and performance evaluation.
  • Optimized and deployed the perception pipeline for real-time inference on Raspberry Pi 4 using OpenVINO for edge-level execution.

January 2025 — October 2025

Research Assistant @ STRESQLab

UNIVERSITY OF WEST ATTICA

  • Contributed to projects involving control systems and artificial intelligence, focusing on system modeling and algorithm implementation.
  • Researched and implemented TinyML solutions on microcontrollers for real-time inference in resource-constrained embedded systems.

October 2024 — June 2025

Autonomous Systems Engineer @ NXP Cup Autonomous Racing

NXP SEMICONDUCTORS

  • Developed a 1:10 scale autonomous racing vehicle using MATLAB/Simulink with a model-based design approach.
  • Designed full system architecture using block diagrams for vehicle dynamics, control systems, and embedded components.
  • Implemented real-time lane-following and autonomous navigation using dual line-scan cameras for track detection and control.

January 2023 — June 2025

Education

August 2026 — August 2028

MEng Systems and Control

Delft University of Technology

NETHERLANDS · GPA 8.0 / 10 (CURRENT)

October 2019 — July 2025

BEng Electrical and Electronics Engineering

University of West Attica

ATHENS, GREECE · GPA 7.43 / 10

Certifications

MathWorks Certified MATLAB Associate Exam

MATHWORKS

MATLAB for the MathWorks Certified MATLAB Associate Exam preparation and certification.

Build MATLAB Proficiency

MATHWORKS

Hands-on MATLAB proficiency track covering data analysis, algorithm development, and simulation workflows.

Python Object Oriented Programming

IEEE SB UNIWA

Object-oriented design principles and patterns applied in Python for scalable software development.

WHAT'S NEXT?

Get In Touch

I'm currently open for new opportunities. Whether you have a question or just want to say hi, I'll try my best to get back to you!

Say Hello
Built with v0