I'm a Robotics & AI Specialist at Atlab in Dubai, UAE, with a background in Automotive Mechatronics Engineering. I build robots end to end (mechanism, electronics, firmware and perception) and I train engineers and educators to do the same.
At Atlab I develop and integrate robotic systems with ROS 2, RViz and Gazebo across robot arms, autonomous vehicles, humanoid and quadruped robots and UAVs, and deploy YOLO/ONNX perception on NVIDIA Jetson AGX Orin.
🌐 Portfolio: abdallahhessam.github.io, with detailed case studies of the projects below.
Technical mentorship for university students and graduate engineers programming Hula drones for autonomous missions.
- Built and taught reference solutions: DFS maze discovery, A* path planning and ONNX (YOLO11) object detection.
- Flown on real drones, using proximity sensors and QR-code positioning.
- Teams I mentored went on to place 2nd and 3rd.
Tech: Python · pyhula SDK · A* / BFS / DFS · ONNX Runtime · YOLO11 · OpenCV · Tkinter
A combat robot built by Team Pegasi, which I founded, for Metal Monsters at the Alamein Robotics Championship in Egypt.
- 4th place at Metal Monsters, 4th edition.
- Led the mechanical design: welded 2 mm steel chassis, chain drivetrain and a 350 W spinner weapon.
- ESP32 control over a Bluetooth PS4 controller, with weapon arming and failsafes.
Tech: ESP32 · PlatformIO · Motor Control · Mechanical Design · Electronics
Autonomous-driving software by the E-Rally team at Helwan University for EVER 2024 in Egypt. I was part of the autonomous-track sub-team.
- ROS 1 driving stack in CoppeliaSim across four tracks, including a custom track.
- YOLO detection with single-camera distance estimation.
- Kalman-filtered odometry and PID control for speed, steering and pedestrian stops.
Tech: ROS 1 · Python · YOLO · OpenCV · CoppeliaSim · MATLAB
A Raspberry Pi field robot that checks plant health, reads the soil, waters, harvests and reports to a server.
- TensorFlow Lite leaf classification (healthy, rust, powdery) with data fusion of soil moisture and pH readings.
- MQTT telemetry to a Mosquitto broker, with SQLite storage and analysis.
- Watering pump and servo harvesting arm.
Tech: Raspberry Pi 4 · Python · TensorFlow Lite · OpenCV · MQTT · SQLite
My graduation project (grade A+): an Adaptive AUTOSAR-inspired communication stack for an infotainment system on Raspberry Pi.
- Distributed communication with SOME/IP over Ethernet, plus MQTT publish/subscribe.
- Service discovery, remote procedure calls and event-based communication.
- Execution management and core types in modern C++.
Tech: C++ · CMake · SOME/IP · Adaptive AUTOSAR · Raspberry Pi · Embedded Linux
- RC Car with STM32: remote-controlled car on an STM32 ARM microcontroller, firmware in C++.
- RGB LED Color Blending: STM32 PWM color mixing, flashed with ST-LINK.
- Robotics: ROS 2, ROS 1, Gazebo, RViz, URDF, autonomous navigation, PID control
- Programming: Python, Modern C++, Embedded C, MATLAB/Simulink, Bash
- AI & Computer Vision: OpenCV, YOLO, ONNX Runtime, TensorFlow Lite, object detection
- Embedded Hardware: NVIDIA Jetson AGX Orin, STM32, ESP32, Arduino, Raspberry Pi, PCB design
- Communication: CAN, LIN, SOME/IP, Ethernet, MQTT, UART, I2C, SPI
- CAD & Fabrication: SolidWorks, Fusion 360, OnShape, 3D printing
- Training & Education: educator training, curriculum development, competition design
- 🤖 ROS 2 Foundations, Packt (Coursera)
- 💡 Embedded C++ Diploma, Robotics Corner
- 🧩 LEGO Education Academy Teacher Trainer
- 🎓 STEM.org Certified Master Trainer
- 🚗 AUTOSAR MasterClass, SPRINTS
- 🥈 2nd place, NASA Space Apps Cairo 2019 (team leader)
- 🔬 National finalist, ISEF Egypt 2017
- Robot learning and embodied AI
- Humanoid and quadruped robotics
- Autonomous navigation and robot perception
- Automotive software and embedded systems
