Dynamic Legged Systems Lab, IIT Genova

Dynamic Legged Systems Lab, IIT Genova Our research lab specializes in the design and control of legged robots for unstructured environment For more information, videos, publications etc.

The Dynamic Legged System lab of the IIT is home to the hydraulic quadruped robots HyQReal, HyQ, HyQ2Max, and MiniHyQ. please visit:
http://dls.iit.it

08/07/2025
08/05/2025

Habemus Paper! Our paper “Multi‑Sensor Fusion for Quadruped Robot State Estimation using Invariant Filtering and Smoothing” has been accepted to IEEE Robotics and Automation Letters! We introduce E‑InEKF and E‑IS, two frameworks that fuse kinematics, IMU, LiDAR odometry, and GPS to reduce position drift, while keeping computation light. Tested on the KAIST HOUND2, we see up to 28% lower trajectory error indoors and 40% outdoors compared to LIO‑SAM and FAST‑LIO2.

This work was carried out in collaboration with our collaborators at the DRCD lab of KAIST.

Authors: Ylenia Nisticò*, Hajun Kim*, João Soares, Geoff Fink, Hae‑Won Park, and Claudio Semini
* Equal Contribution
Links to the paper:
1) Arxiv: https://www.arxiv.org/abs/2504.20615
2) IEEE Xplore: https://ieeexplore.ieee.org/document/10977997

DLS  2025: We were honoured to present our Vinum project at Euroflora 2025 in Genoa, showcasing advancements in automate...
02/05/2025

DLS 2025: We were honoured to present our Vinum project at Euroflora 2025 in Genoa, showcasing advancements in automated winter pruning technology.

Developed by the ISTITUTO ITALIANO DI TECNOLOGIA (IIT) and Università Cattolica del Sacro Cuore, and supported by the innovation ecosystem, Vinum represents a significant step forward in precision agriculture.

Claudio Semini and Angelo Bratta gave an overview of the challenges that agriculture is facing nowadays, highlighting how Vinum and robotics in general can help to solve them.

After that, visitors experienced live demonstrations of the system in action.

Proud to be part of Robot Valley 2025 at Villa Durazzo Bombrini, organized by RAISE Liguria! 🤖Yesterday, today, and tomo...
11/04/2025

Proud to be part of Robot Valley 2025 at Villa Durazzo Bombrini, organized by RAISE Liguria! 🤖

Yesterday, today, and tomorrow (April 10th, 11th, and 12th), our lab is showcasing VERO—the quadruped robot powered by AI to help remove cigarette butts littered in the environment.

This morning, our PI, Claudio Semini, took part in a round table discussion on robotics and AI with other experts. It’s been a fantastic opportunity to exchange ideas and shape the future of innovation together 💡.

Check out the articles for more details:
- Robot Valley 2025 (https://www.raiseliguria.it/2025/04/10/genova-apre-le-porte-a-robot-valley-2025-arte-tecnologia-e-creativita-al-centro-della-seconda-edizione/)
- VERO, our AI-driven cigarette butt removal robot (https://www.raiseliguria.it/2024/07/03/robot-quadrupede-vero-guidato-da-intelligenza-artificiale-elimina-mozziconi-di-sigaretta/)

Stay tuned for more from Robot Valley 2025 and our ongoing efforts to make robots an integral part of tomorrow’s sustainable world! 🌍

ISTITUTO ITALIANO DI TECNOLOGIA

27/03/2025

Our paper “MUSE: a Real-Time Multi‑sensor State Estimator for Quadruped Robots” has been accepted for publication in IEEE Robotics and Automation Letters (RA-L)! In it, we introduce MUSE, a novel sensor‑fusion state estimator that tightly integrates onboard IMU, encoder, camera, and LiDAR data to deliver accurate, real‑time state estimation for quadruped robot navigation. MUSE has been demonstrated in closed‑loop experiments and enables robust autonomous locomotion even on slippery, uneven terrain. By fusing complementary sensors, MUSE improves localization accuracy, reducing error by up to 67.6% compared to leading methods.

Authors: Ylenia Nisticò, João Carlos Virgolino Soares, Lorenzo Amatucci, Geoff Fink, Claudio Semini
Paper: https://arxiv.org/abs/2503.12101
Code: https://github.com/iit-DLSLab/muse/tree/main

We are thrilled to announce our latest research, "SANDRO: a Robust Solver with a Splitting Strategy for Point Cloud Regi...
19/03/2025

We are thrilled to announce our latest research, "SANDRO: a Robust Solver with a Splitting Strategy for Point Cloud Registration", has been accepted to the 2025 IEEE International Conference on Robotics and Automation (ICRA). In this work, we introduce SANDRO, an innovative algorithm that enhances point cloud registration by combining an Iteratively Reweighted Least Squares (IRLS) framework with a robust loss function and a novel splitting strategy. This approach effectively addresses high outlier rates and skewed outlier distributions, leading to significant improvements in registration success rates. Notably, SANDRO demonstrates a 20% improvement over current state-of-the-art methods on the Redwood real dataset and a 60% improvement on synthetic data. We invite you to explore our findings in detail in the full paper.

Authors: Michael Adlerstein, Joao Carlos Virgolino Soares, Angelo Bratta, Claudio Semini

Read the full paper here: https://arxiv.org/pdf/2503.07743v1
Our code will be available at https://github.com/iit-DLSLab/SANDRO during the ICRA 2025 conference!

Congratulations to our new Doctor Ylenia Nisticò who successfully discussed her thesis titled "Enhancing State Estimatio...
22/02/2025

Congratulations to our new Doctor Ylenia Nisticò who successfully discussed her thesis titled "Enhancing State Estimation in Quadruped Robots: Multi-Sensor Fusion for Challenging Terrains", at IIT Genova last Thursday.
Congratulations also to the supervisors João Carlos Virgolino Soares, Geoff Fink, and Claudio Semini.

You can find out more about Ylenia's research and other interesting works, on our website (https://dls.iit.it/).

Check out our dual-arm HyQReal robot as part of our Inail Robot Teleoperativo project in collaboration with Inail and th...
11/02/2025

Check out our dual-arm HyQReal robot as part of our Inail Robot Teleoperativo project in collaboration with Inail and the ISTITUTO ITALIANO DI TECNOLOGIA research units of Nikos Tsagarakis with Ioannis Sarakoglou, PhD, Siddharth Deore and Darwin Caldwell with Yonas Teodros Tefera and our team Dynamic Legged Systems with Claudio Semini, Victor Barasuol, Matteo Villa and many others!
IEEE Spectrum just published a great article about our most recent results: https://spectrum.ieee.org/telepresence-robot

IIT’s hydraulic quadruped can carry a pair of massive arms

03/01/2025

The combinatorial nature of contact planning in legged robots hinders the success of optimal control in navigating complex scenarios. In our recent IEEE RA-L paper, we demonstrate how sampling-based methods and supervised learning techniques can be coupled to search for a solution to this problem in real-time.

Title: Non-Gaited Legged Locomotion with Monte-Carlo Tree Search and Supervised Learning
Authors: Ilyass Taouil*, Lorenzo Amatucci*, Majid Khadiv, Angela Dai, Victor Barasuol, Giulio Turrisi, Claudio Semini
*Equal Contribution

Link to the paper: https://arxiv.org/abs/2408.07508

ISTITUTO ITALIANO DI TECNOLOGIA

VERO at Genova Smart Week 2024!Today we brought our VERO robot to the Genova Smart Week 2024! The Municipality of Genoa ...
29/11/2024

VERO at Genova Smart Week 2024!

Today we brought our VERO robot to the Genova Smart Week 2024! The Municipality of Genoa presented the most innovative projects aimed at improving residents' quality of life. VERO, our walking vacuum-cleaner robot, is taking part in the project “il mare comincia qui” led by the International School of Genoa, in collaboration with IIT, UniGenoa, Out Be, and Re-Cig, that fights the littering of cigarette filters that destroy the environment.
This event was a great opportunity to showcase how VERO can contribute to a more efficient and sustainable future for cities.

ISTITUTO ITALIANO DI TECNOLOGIA

Indirizzo

Via San Quirico, 19D
Sant'olcese

Notifiche

Lasciando la tua email puoi essere il primo a sapere quando Dynamic Legged Systems Lab, IIT Genova pubblica notizie e promozioni. Il tuo indirizzo email non verrà utilizzato per nessun altro scopo e potrai annullare l'iscrizione in qualsiasi momento.

Scelte rapide

Condividi