I build autonomous robotic systems that combine perception, planning, learning, and real-time control, with a focus on aerial robotics, computer vision, reinforcement learning, and navigation in constrained environments.
-
Damage-Robust Quadruped Locomotion - PPO locomotion controller for the MuJoCo Ant, trained under randomized single-leg actuator degradation to improve performance after damage.
Tech: Python · reinforcement learning · PPO · Stable-Baselines3 · Gymnasium/MuJoCo - Repository -
SimBioSys Greenhouse Robot - Autonomous ROS2 greenhouse robot for flower-bed inspection, navigation, UI interaction, and basic harvesting.
Tech: ROS2 · Nav2/AMCL/SLAM · YOLO perception - Repository -
Parrot Bebop Obstacle Avoidance - Onboard perception and reactive obstacle-avoidance pipeline for a physical Parrot Bebop drone using Paparazzi UAV.
Tech: C · Paparazzi UAV · onboard computer vision - Repository -
RRT-MPC Drone Navigation - Autonomous quadrotor navigation in simulation using sampling-based planning and model predictive control.
Tech: Python · RRT/RRT* · MPC - Repository -
Multi-Sensor Pedestrian Detection - Autonomous-driving perception pipeline for pedestrian detection and 3D localization using camera, LiDAR, and radar data.
Tech: Python · sensor fusion · CNN/NMS - Repository -
Duck Collector Imitation Learning - Vision-based imitation-learning policy for collecting objects in MiniWorld environments.
Tech: Python · behavior cloning · Keras - Repository -
Hannes Hand ROS Evaluation - ROS-based experimental framework for evaluating control strategies on the Hannes prosthetic hand.
Tech: ROS · Python · prosthetic-hand control - Repository -
Single-Cycle RISC-V CPU - Single-cycle RISC-V CPU implemented in Verilog and verified through waveform-based simulation. Tech: Verilog · RISC-V · GTKWave - Repository
-
FPGA HDMI Display Controller - Custom Verilog HDMI display peripheral connected to a Hummingbird/e203 CPU through memory-mapped I/O.
Tech: Verilog · FPGA · HDMI - Repository
- Robotics and autonomous systems
- Reinforcement learning and imitation learning
- Robust control and policy evaluation
- Motion planning and trajectory optimization
- Computer vision and sensor fusion
- Embedded systems and real-time software
- Python, C/C++, ROS2, PyTorch, OpenCV, Stable-Baselines3, Gymnasium, MuJoCo, and Verilog
LinkedIn: tommaso-calzolari
Email: tommi.calzolari@gmail.com
TU Delft: tcalzolari@tudelft.nl