Building the open-source software stack for intelligent robotics

MAESTRO brings together leading research institutions and industrial partners to build and sustain an open-source software stack enabling robots to move, manipulate and interact reliably in complex, real-world environments.

Our ambition

Robots are moving beyond controlled environments and into increasingly complex, dynamic and human-centered settings.

To operate reliably in the real world, they need software capable of modelling physical interactions, anticipating motion, handling contacts and optimizing their actions, all while meeting the performance requirements of real robotic systems.

MAESTRO brings these capabilities together into a shared, open-source software stack. Our ambition is to build a robust and sustainable European software foundation for intelligent robotics: open to researchers, relevant to industry, and designed to accelerate the transfer of innovation from research to real-world applications.

The software stack

The MAESTRO stack brings together software developed over the years by French research institutes, namely Inria and CNRS, through significant and sustained efforts in both research and engineering. Research activities involve scientists, PhD students and postdoctoral researchers advancing the state of the art in robotics, optimization, simulation and control. Engineering teams turn these advances into robust, reusable software, supported by robotic experimental platforms where new methods and technologies can be tested and validated.

All the software components currently envisioned in the Consortium have been and are currently developed by Inria and CNRS under permissive Open Source Software licenses, mostly BSD.

Together, they cover key capabilities in robot modelling, simulation, optimization and control:

• Pinocchio - Robot dynamics
• Coal - Collision detection
• Aligator - Optimal control
• ProxSuite - Numerical optimization
• Simple - Differentiable physics simulation
• EigenPy - Python bindings for Eigen

The components are developed as an integrated software ecosystem, with shared engineering standards, continuous integration and a common technological roadmap.

Research and industry, building together

MAESTRO brings together research organizations and industrial partners around a shared challenge: turning advances in robotics and Physical AI into robust, reusable technologies.

Research partners

Inria · CNRS

Industrial partners

Renault · Safran · Wandercraft · Helsing · Atos

Industrial partners bring real-world use cases, constraints and requirements, helping shape the Consortium’s technological roadmap and accelerate the transition from research to real-world deployment.

This collaboration creates a direct bridge between scientific research, software engineering, experimental validation and industrial applications.

What the Consortium does

Maintain

Ensure the long-term maintenance, quality and sustainability of the core software components.

** Integrate**

Turn advances from research into robust software that can be used by the wider robotics community.

Standardize

Develop shared standards, benchmarks and engineering practices across the software stack.

Connect

Bring researchers, engineers and industrial users together around a common technological roadmap

Sustain

Build the governance, engineering capacity and community needed to make the MAESTRO software ecosystem sustainable over the long term.

An open consortium built for the long term

MAESTRO is built around an open consortium model designed to preserve and grow a shared technological commons.

Research institutions provide the scientific and engineering foundations of the software stack, while industrial partners help shape its roadmap around real-world needs and can contribute to its development.

Under Inria’s institutional framework, this model aims to ensure the long-term sustainability and independence of the MAESTRO ecosystem while keeping its software open and accessible to the wider robotics community.

Want to join the MAESTRO community?