About
Michele Focchi is a world-recognized expert in motion planning and control of quadruped robots, with 16 years of experience in robotics. He is particularly known for his pioneering work on heuristic locomotion in unstructured terrains. His research lies at the intersection of control, optimization, and machine learning, with a focus on enhancing quadruped robot performance in challenging environments through optimization-based techniques. Currently, he is a Scientific Advisor for All3. From 2022 to 2024, he was a Researcher at the University of Trento in the Department of Information Engineering and Computer Science (DISI), where he taught introductory robotics courses. He was also a visiting scientist at the Istituto Italiano di Tecnologia (IIT), where previously he co-founded the Dynamic Legged Systems (DLS) Lab, a leading international research team dedicated to the development of quadruped robots and their locomotion. Dr. Focchi earned both his B.Sc. and M.Sc. in Control Systems Engineering from Politecnico di Milano. In 2013, he completed his Ph.D. in Robotics at IIT, contributing to the Hydraulically Actuated Quadruped Robot project. His work evolved from low-level controllers for locomotion to whole-body control, model identification, and robust locomotion strategies for real-world platforms. He has also explored innovative robotic platforms, including a rappelling robot for hydro-geological risk reduction and control strategies for tracked robots in agriculture.
Beyond academia, Dr. Focchi has played key roles in several high-profile industrial and academic projects, including ECHORD++, INAIL, and ANT with the European Space Agency. In 2015, he co-founded the MOOG-IIT joint lab to develop advanced software and control solutions for autonomous robots. He has organized several scientific workshops and delivered over 20 invited talks at international events.
He is an Associate Editor for RA-L and the ICRA conference. He has authored or co-authored more than 53 scientific papers in international journals and conferences, with a high citation record, and has supervised numerous Master’s and Ph.D. theses.
Previous research works (in chronological order):
You can check more about these in my Youtube Research playlist
-
Digital agriculture
-
Climbing robots
-
Space exploration with legged robots
-
Model predictive control applied to locomotion of legged robots
-
Dynamic planning for aerial motions
-
Heuristic planning on rough terrain for quadruped locomotion
-
Development of reactive modules for locomotion (e.g. step reflex, slip recovery algorithms)
-
Model identification
-
Whole-body control
-
Torque control
Projects I was involved:
-
01/2021 - 31/12/2022 European Space Agency - Autonomous Non-Wheeled all-Terrain Rover(ANT) - 400keuro This project is a joint work carried out by the German Research Center for Artificial Intelligence (DFKI), IIT and Airbus Defence and Space Ltd. (ADS). The objective of the activity is to develop the ANT navigation system for legged robots. It will be able to perceive the terrain, to plan a path to a desired 3goal and to control the path execution while traversing unconsolidated, inclined, and rugged terrain. A modular generic approach is being developed to exploit the potential of robots with four (quadrupeds) as well as with six legs (hexapods).
-
09/2017 - 31/12/2022 INAIL - The Teleoperazione project - 5.4 Meuro. This project aims to enhance occupational safety in hazardous environments through substitution, i.e., removing the worker from the unsafe area and using robotic technologies to do the same tasks through remote robotic teleoperation. The project proposes a collaborative system: a centaur-like robot featuring a hydraulically-driven quadruped platform, a dexter- ous manipulator arm, and advanced 3D user interfaces designed for intuitive remote teleoperation. The manipulator will have enough degrees of freedom to ensure dexterity. The human-robot interface will be improved with augmented reality with tactile feedback for tele-operation.
-
09/2016 - 07/2018 HyQ-REAL (EU funded FP7 ECHORD++ experiment- 300keuro) Bring HyQ from the research lab to the real-world application. Developing planning algorithms for loco- motion on unstructured terrain.