Machine Intelligence for Robot Autonomy Lab
Welcome to the official website of the MIRA Lab!
Our goal is to develop the algorithmic foundations and system-level methodologies that enable AI-powered autonomous systems to operate safely, efficiently, and reliably in high-stakes, real-world environments. We work on employing and advancing techniques from AI / ML, control theory, and mathematical optimization — and apply our results to aerospace robotics, future mobility systems, and autonomy at large.
We are part of the Italian Institute of Artificial Intelligence (AI4I), hosted at Officine Grandi Riparazioni (OGR) in the beautiful city of Turin.
>
Research Thrusts
01
Robot Learning
We build tools for Physical AI that make generalist robot policies reliably grounded in the physical world, developing novel learning algorithms, reasoning mechanisms, and approaches that combine learning with model-based methods. These tools are designed to run in real time on real hardware and to generalize across tasks, environments, and robot platforms.
02
Coordinated Autonomy
As autonomous systems become increasingly central to society, they will rarely act alone: fleets of vehicles moving people through a city, constellations of spacecraft sharing an observation task, teams of robots operating alongside each other and alongside humans. We develop learning and optimization methods for decision-making across many interacting agents and connect individual behavior to system-level outcomes.
03
Physical AI Safety
Acting in the physical world introduces a category of risk that digital systems do not face: decisions are made under real-time constraints, in dynamic environments, and mistakes have physical consequences. We develop methods to ensure that autonomous systems operate safely and reliably through runtime monitoring of system behavior, guardrailing of learned policies, and reasoning-based mechanisms that anticipate novel hazards and synthesize recovery behavior on the fly.
04
Societally-impactful Autonomy
Ultimately, the goal of our lab is to push the boundaries of autonomy in high-stakes, safety-critical domains. These environments force existing methods to operate at the edge of what is currently possible, accelerating the next generation of autonomous systems: “a wind tunnel for Physical AI”.
>
Latest News
Jun 2026
Our paper SAGES: Semantic Trajectory Generation for Goal-Oriented Spacecraft Rendezvous was selected as the Intelligent Systems Best Paper Award at the 2026 AIAA SciTech Forum!
Apr 2026
The Principles of Robot Autonomy book is now complete and available online!
Mar 2026
Our work on Multi-Timescale Model Predictive Control for Slow-Fast Systems was accepted at ECC 2026!
Mar 2026
Our paper Graph Neural Model Predictive Control for High-Dimensional Systems was accepted at ICRA 2026! Introducing scalable and effective learning-based MPC for soft robotics.
Mar 2026
Our paper Reproducibility in the control of autonomous mobility-on-demand systems was published in IEEE T-RO!
Mar 2026
Another year of AA203: Optimal and Learning-based Control at Stanford begins!
Feb 2026
Two IEEE RA-L papers accepted on learning-based control of multi-agent robotic systems: GRAND: Guidance, Rebalancing, and Assignment for Networked Dispatch in Multi-Agent Path Finding and Accelerating High-Capacity Ridepooling in Robo-Taxi Systems .
Jan 2026
Very happy to host the 4th Workshop on Artificial Intelligence for Space at CVPR 2026! See the full call for papers .
Nov 2025
Wonderful visit at NASA JPL ! I gave a talk on "Advancing Aerospace Autonomy with Foundation Models".
Jul 2025
Our work on multi-modal transformers for spacecraft trajectory optimization was selected as single-track presentation at ACC 2025.
May 2025
Our works on vision foundation models for anomaly detection and transformers for model predictive control were presented at ICRA 2025.
Apr 2025
Our ICLR paper on Offline Hierarchical Reinforcement Learning via Inverse Optimization was presented in Singapore.
Apr 2025
Another year of AA203: Optimal and Learning-based Control at Stanford begins!
Mar 2025
Check out our two IEEE Aerospace Conference papers: Space-LLaVA: a Vision-Language Model Adapted to Extraterrestrial Applications and Towards Robust Spacecraft Trajectory Optimization via Transformers .
Mar 2025
New pre-print: Robo-taxi fleet coordination at scale via reinforcement learning .
Selected Press Coverage
>
Featured Research
Space-LLaVA: a Vision-Language Model Adapted to Extraterrestrial Applications
M. Foutter, D. Gammelli, J. Kruger, E. Foss, P. Bhoj, T. Guffanti, S. D'Amico, M. Pavone
IEEE Aerospace 2025