Emanuele Cuzzocrea
I am currently a first year PhD student in Robotics and Intelligent Machines at the Italian Institute of Technology (IIT) in Genoa, within the Humanoid & Human Centered Mechatronics (HHCM) Lab, under the supervision of Dr. Nikolaos Tsagarakis. My research focuses on the application of foundation models (i.e., LLM, VLM, VLA) to legged and hybrid wheeled-legged robots, with particular emphasis on perception-driven loco-manipulation and navigation tasks.
During my Master’s studies, I conducted research at the PRISMA Lab in Naples, where I contributed to the hardware construction of the SOLO 12 quadruped robot and carried out my thesis, titled “Analysis of Quadruped Robot Gaits in Push-and-Slide Interaction Tasks”, supervised by Prof. Fabio Ruggiero, conducted on the ANYmal-D quadruped robot by ANYbotics and submitted to IEEE Robotics and Automation Letters (RA-L) in January 2026.
My research interests include legged robots, foundation models, and robot perception. Outside of robotics, I enjoy speedcubing, tennis, golf, chess, drawing, and video game development.
Publications
Selected Course Projects
MPC vs Deep RL for Autonomous Quadruped Locomotion
Comparison of model-based and data-driven control strategies for quadruped robots, including MPC and Deep Reinforcement Learning with Reward Machines. Evaluation of multiple gaits and planning-aware locomotion with RRT-based obstacle avoidance. Simulations in ROS (Gazebo, RViz) and RaiSim using Unitree A1 and Aliengo platforms.
Learning-Based Gait Transitions for Quadruped Robots
Autonomous framework for dynamic gait transitions in quadruped robots using Deep RL with Reward Machines and high-level decision-making for autonomous gait switching. Trained and validated in RaiSim on the ANYmal-B platform.
Virtual Prototyping of Nuclear Remote Handling Tools
Virtual prototyping of an auxiliary handling system for a nuclear fusion reactor, developed in CATIA. Modular design integrating a KUKA manipulator, scissor lift, and omnidirectional cart, validated through FEM, kinematic, and ergonomic (Jack) analyses.
Perception-Aware Navigation and SLAM for Mobile Robots
Design and implementation of a mobile robot navigation pipeline in ROS, including trajectory tracking, goal-based navigation, mapping and SLAM, and vision-based tasks. Development and validation performed in custom Gazebo environments.
Experience
Education
Hobby
Rubik's Cube in under 9 Seconds
Speedcuber VS Robot
Home Made Humanoid Robot
Video Game Development