Bachelor’s Degree in Mechatronics Engineering
Fırat University · Elazığ, Türkiye
Academic & Professional Portfolio
I build autonomous robots and the embedded systems behind them.
Robotics and embedded systems engineer focused on autonomous systems. I integrate advanced sensing, SLAM, and AI to build dependable robotic systems for real-world environments.
0Projects
0Publications
0Skills
PresentingRescue-robot payload · Creativity & Innovation Conference, Oct 2026
IN PICTURES
The work, up close.
01 / EDUCATION
Academic foundation
02 / PUBLICATIONS
Research & publications
Drift-Resilient Indoor 3D Mapping in GPS-Denied Environments Using RKO-LIO with Velodyne–Pixhawk Integration in ROS 2
A drift-resilient LiDAR–IMU mapping system for GPS-denied indoor environments, integrating RKO-LIO with a Velodyne VLP-16 and Pixhawk Cube Orange+ under ROS 2.
Read publicationPV-STAM: Velocity-Aware Attention for Mapless Deep Reinforcement Learning Navigation in Dynamic Environments
A compact attention module that lets a robot with only a 2D LiDAR tell approaching obstacles from static ones, trained with deep reinforcement learning and validated in 130 physical TurtleBot3 trials.
Read publicationMSR-SLAM: A Multi-Stage Map-Refinement Framework for Reliable LiDAR–Inertial SLAM in GPS-Denied Environments
A map-refinement back end for LiDAR–inertial SLAM that decides which loop closures deserve to change the map, using geometric verification and an online admission check, tested on a Unitree Go2 quadruped and KITTI.
Read publicationDemonstrating the Efficiency of RKO-LIO on a DJI UAV Platform Utilizing Drone Built-in IMU for GPS-Denied Mapping
A hardware and software framework that integrates a DJI Matrice M350 RTK with an NVIDIA Jetson Orin NX over ROS 2, using only a Velodyne VLP-16 and the drone's built-in 200 Hz IMU for GPS-denied LiDAR-inertial mapping with RKO-LIO.
Read publication03 / EXPERIENCE
Work experience
Volunteer Researcher
RAI Laboratory, Fırat University
Research and development for autonomous disaster management using heterogeneous robotic systems within the TÜBİTAK 1001 project.
Robotics Research Intern
RAI Laboratory, Fırat University
Supported hardware-software integration of the DJI Matrice 350 RTK with NVIDIA Jetson Orin NX.
04 / TECHNICAL CAPABILITIES
Skills, organised by system layer.
61 tools and methods across 9 engineering disciplines.
01
6 skillsAutonomy & Navigation
02
7 skillsControl & Simulation
03
6 skillsElectronics & PCB
04
8 skillsEmbedded & Edge AI
05
6 skillsFabrication & Testing
06
8 skillsMechanical Design & CAD
07
7 skillsPerception & AI
08
5 skillsRobotics Middleware
09
8 skillsSoftware & Data
05 / PROJECT ARCHIVE
Selected work, fully documented.
View case study →Research prototype
Offline AI Payload for Search & Rescue Robots
A modular payload for rescue robots that combines thermal and visual detection, a radar sensor, gas and flame sensing, an offline voice assistant and long-range alerts, all running on a small Jetson TX2 computer without internet.
Works fully offline: detection, voice dialogue and alerts need no internet
View case study →Research prototype
Demonstrating the Efficiency of RKO-LIO on a DJI UAV Platform Utilizing Drone Built-in IMU for GPS-Denied Mapping
A mapping payload for the DJI Matrice M350 RTK drone that builds accurate 3D maps of outdoor areas without relying on GPS. A LiDAR scanner and a small onboard computer reuse the drone's built-in motion sensor, so no extra sensor is needed.
Accurate 3D map of a campus area, built entirely onboard the drone
View case study →Simulation project
Autonomous Cave Exploration & Obstacle-Avoidance Drone
A simulated drone that explores caves, tunnels and mines without GPS. It builds a live 3D map from a LiDAR, plans collision-free routes through it, picks its own next target and recovers when it gets stuck.
Builds a live 3D map of GPS-denied tunnels from LiDAR data
View case study →Completed
Robotic Arm: SolidWorks Design to Simscape Control
A robotic arm taken from CAD design to physics simulation. SolidWorks models are imported into Simscape Multibody, inverse kinematics turn a target point into joint angles, and PID control is demonstrated on a simpler two-joint version.
Connected SolidWorks geometry and mass properties to a Simscape physics model
View case study →Completed
Automated Coffee Preparation System
An embedded automation prototype that coordinates preparation stages from a Raspberry Pi control core. Modular Python software, relay-switched loads, PWM servo actuation, and a local LCD/keypad interface were brought together to automate ingredient dispensing, pumping, and heating-related control.
Connected AC and DC actuator stages to one Raspberry Pi control workflow
View case study →Completed
CNC Multi-Function Conversion System
A multi-function digital-fabrication platform created by converting a 3D printer into an interchangeable-tool CNC system. The documented workflow combines mechanical redesign, SVG-to-G-code preparation, and material-aware calibration for wood carving, plastic engraving, and 2D drawing.
Reconfigured a 3D-printer platform as an interchangeable-tool fabrication system