Control
Model-based controller design, from the linearised model to the tuned controller on the robot.
// robotics · control · embedded systems
Control Engineer at Hilti · M.Sc. Mechatronics & Robotics (TUM)
I take controllers from simulation to reliable real-time code on resource-constrained hardware.
1 patent filed 1 IEEE paper

P-10MicroDuck RL
P-09 · 2026Smallest Variometer
P-08Two-Axis Robotic Platform
P-07 · 20243D Printed Drone
P-06 · 2023Balancing Robot
P-05 · 2023Via Eureka
P-04 · 2023Torque-Sensing Actuator
P-03 · 2023Bilateral Teleoperation
P-02 · 20226 DOF Robotic Arm
P-01 · 2020VariometerSelect a hexagon to open or locate a project
about.md
Hey! I'm Luca Obwegs from Italy. I love robots, climbing and learning new things.
I work as a control engineer in the Mechatronics and Robotics Systems Laboratory at Hilti in Liechtenstein, where I develop control software for robotic and mechatronic prototypes – from modelling and simulation to the implementation on the target hardware and testing on the real system.
In my spare time I build robots, drones and flight instruments from scratch – from the CAD model and the PCB to the embedded software and the control algorithms. What I enjoy most is the step from a controller that works in simulation to one that runs reliably on real, resource-constrained hardware.
Feel free to get in touch or take a look at my work below.
languages
01 // career & education
Mechanical-engineering high school diploma, hands-on electrical assembly and soldering, plus summer internships at M2 Railtech working on mechatronic prototypes, microcontrollers and PCB design.
02 // skills
Model-based controller design, from the linearised model to the tuned controller on the robot.
Fusing noisy sensors into states a controller can rely on.
Control code that is readable, tested and runs on the real system.
Real-time control loops on microcontrollers, close to the sensors and actuators.
Modelling mechanisms and testing controllers before they touch the hardware.
Building the prototypes myself, so the software has something to run on.
03 // personal projects
The projects in this section are personal projects. University coursework is listed separately below; professional projects are not included.

An upcoming build of the open-source MicroDuck biped, focused on learning reinforcement learning, locomotion training and the path from simulation to real hardware.
View project
The successor of my first variometer: a complete flight instrument on a custom STM32 PCB that weighs about 5.5 g including battery and case.
View project
A two-axis tracking prototype and simulation with detailed stepper-servo modelling, sensor fusion and delayed-measurement compensation.
View project
A 3D-printed quadrotor built from scratch: Simulink simulation, custom STM32 flight software with MEKF state estimation, and PID / LQR / MPC control.
View project
A two-wheeled self-balancing robot for comparing PID and LQR control on an STM32 – from the pendulum model to the tuned controller on the robot. MPC is next.
View project
A new 8-pitch, 230 m multi-pitch climbing route on the Lagazuoi above Passo Falzarego (Dolomites), opened together with Lucas in autumn 2023.
View project
A custom robot actuator that measures torque as well as position and velocity – the building block for a force-controlled 3-DOF robot.
View project
Stable bilateral teleoperation of Phantom robots under time delay using passivity-based force and position control – developed at KAIST, South Korea.
View project
A fully 3D-printed six-axis robot arm inspired by the ABB IRB 1100 – stepper drives, numerical inverse kinematics, a teach-in GUI and ROS 2.
View project
A paragliding variometer built from scratch around an ESP8266, with GPS, barometer, accelerometer and Kalman-filter sensor fusion.
View project04 // publications & IP
S. Groß, L. Obwegs, J. Li, T. Spiegeler Castañeda, A. Ganguly, C. Piazza, and S. Haddadin, Transforming Tactile Interfaces: Tri-Axis Force Sensor for Sensory Signal Processing, accepted at IEEE World Haptics 2023, Delft.
Stop Assist for SID – data-driven automated stopping controller for assisting operators in driving screws. Patent filed; 50% contribution.
05 // university coursework
Selected hands-on projects completed as part of robotics and autonomous-systems courses.
A ROS-based differential-drive robot that mapped an unknown environment, found a visual goal, collected a magnetic red ball and delivered it autonomously.
A simulated autonomous drone that used its camera to map a Unity environment and navigate to a goal through a ROS perception, planning and control pipeline.