Available for robotics roles

Can Keçilioğlu

Building perception systems

Final-year mechatronics engineer focused on the software and firmware that lets robots sense, estimate and act — from bare-metal sensor drivers to the state estimation a system can plan on.

View work Get in touch

perception/ekf_node.cpp

// raw sensor → state → world model
while (rclcpp::ok()) {
  auto imu = imu_sub.read();   // 200 Hz, noisy
  ekf.predict(imu, dt);
  if (auto gps = gps_sub.poll())
    ekf.update(*gps);        // 10 Hz, absolute
  publish(ekf.state());        // a pose you can trust
}

Selected work

All work →

01Flagship

Ground-analysis rover

Proprioceptive terrain mapping. IMU + GPS fused through an EKF into a live slope-and-roughness heatmap, with micro-ROS bridging an ESP32 and a Raspberry Pi into a real-time dashboard.

ROS2
micro-ROS
EKF
ESP32
Raspberry Pi
C/C++

02In progress

A from-scratch mechatronic build. The hard part isn't the look — it's making an open-loop stepper behave like a closed-loop system: re-homing every move via a Hall sensor, per-unit EEPROM calibration, custom firmware.

Steppers
Hall sensor
EEPROM
Custom firmware
C/C++

What I do

Perception & sensor fusion

Turning noisy IMU, GPS and range data into clean, trustworthy state — with filters like the EKF at the core.

Embedded & firmware

Bare-metal C/C++ on microcontrollers, deterministic sampling, and bridging hardware into ROS2 over micro-ROS.

Systems & integration

Designing the whole path — sensor to estimate to world model — and making the seams between them reliable.

Let’s build something that perceives.

Open to robotics and embedded roles, and to hard problems in perception and real-time systems.

Get in touch