Andrea Patrizi

Andrea Patrizi

Auto-generated profile, based on publicly available information

Andrea Patrizi

Humanoids and Human-Centered Mechatronics Research Line, Istituto Italiano di Tecnologia (IIT), Genova, Italy

Research Focus & Metrics

Academic Career Length

4Years (2023-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
2 April 2025 in IEEE Robotics and Automation Letters
Verified
Journal Article
Closed access
RDRui Dai
Rui Dai
LRLuca Rossini
Luca Rossini
ALArturo Laurenzi
Arturo Laurenzi
APAndrea Patrizi
Andrea Patrizi
NTNikos Tsagarakis
Nikos Tsagarakis
The complex nonlinear nature of friction in real-world applications renders traditional physical models inadequate for accurately capturing its characteristics. While numerous learning-based approaches have addressed this challenge, they often lack interpretability and fail to uphold the physical guarantees essential for reliable modeling. Additionally, existing structured data-driven methods, despite their efficacy in handling nonlinear systems, seldom account for the specific traits of friction or ensure passivity. To overcome these limitations, we introduce a structured Gaussian Process (GP) model that adheres to the physical consistency of joint friction torque, enabling data-driven modeling in function space that accurately captures Coulomb and viscous friction characteristics while further guaranteeing passivity. We experimentally validate our approach by deploying the friction model on a two-degree-of-freedom (2-DoF) leg prototype. Our approach exhibits robust performance in the presence of non-passive and high-noise data. Experimental results demonstrate that our joint friction model achieves enhanced data efficiency, superior friction compensation performance, and improved trajectory tracking dynamics compared to other friction models.

19 March 2026 in IEEE Robotics and Automation Letters
Verified
Journal Article
Closed access
APAndrea Patrizi
Andrea Patrizi
CRCarlo Rizzardo
Carlo Rizzardo
ALArturo Laurenzi
Arturo Laurenzi
FRFrancesco Ruscelli
Francesco Ruscelli
LRLuca Rossini
Luca Rossini
NTNikos G. Tsagarakis
Nikos G. Tsagarakis
We propose a contact-explicit hierarchical architecture coupling Reinforcement Learning (RL) and Model Predictive Control (MPC), where a high-level RL agent provides gait and navigation commands to a low-level locomotion MPC. This offloads the combinatorial burden of contact timing from the MPC by learning acyclic gaits through trial and error in simulation. We show that only a minimal set of rewards and limited tuning are required to obtain effective policies. We validate the architecture in simulation across robotic platforms spanning $\mathbf {50}\,$kg to $\mathbf {120}\,$kg and different MPC implementations, observing the emergence of acyclic gaits and timing adaptations in flat-terrain legged and hybrid locomotion, and further demonstrating extensibility to non-flat terrains. Across all platforms, we achieve zero-shot sim-to-sim transfer without domain randomization, and we further demonstrate zero-shot sim-to-real transfer without domain randomization on Centauro, our $\mathbf {120}\,$kg wheeled-legged humanoid robot.

12 December 2023
Verified
Conference Paper
Closed access
APAndrea Patrizi
Andrea Patrizi
FRFrancesco Ruscelli
Francesco Ruscelli
ALArturo Laurenzi
Arturo Laurenzi
NTNikos G. Tsagarakis
Nikos G. Tsagarakis
In this work we showcase a comprehensive study, developed around our heavy-duty electrically-actuated quadruped leg prototype, focusing on the task of jumping. Our study employs a two-stage pipeline for the design and execution of agile and efficient jumping maneuvers. We combine offline non-linear whole-body Trajectory Optimization (TO) to formulate and solve two distinct non-linear problems, each tackling different phases and performance aspects of the jumping maneuver. The first one jointly addresses the thrust and flight phases and aims at maximizing the achieved jump height, while accounting for several hardware limitations. The second problem deals with the impact and braking phases minimizing the kinetic energy dissipated at the impact, while maximizing the energy regenerated by the actuators during braking. We validate the accuracy of the employed models and the results of the optimizations both in simulation and on the real leg hardware, showing the effectiveness of our approach in maximizing hardware performance and the benefits of exploiting combined impact-energy considerations during the execution of jumping maneuvers.