Code

dm_robotics

dm_robotics is a collection of libraries, tools, and tasks developed by Google DeepMind specifically for robotics research. It provides a comprehensive framework for building and experimenting with robotic agents.

Key components include:

• AgentFlow: A library for composing reinforcement learning agents and distinct sub-tasks.
• MoMa (Mobile Manipulation): A detailed framework for defining manipulation tasks and environments, supporting both simulated (via MuJoCo) and real-world robot setups.
• Geometry & Transformations: Primitives for handling scene geometry and rigid body transformations.
• Vision: Tools for visual blob detection and tracking.
• Manipulation: “RGB” object meshes and utilities for creating simulated robotic cells.
• Controllers: Robust controllers including QP-optimization based Cartesian controllers.

This library is often used in conjunction with dm_control to create advanced physics-based simulation environments.

YARP

YARP is a library and toolkit for communication and device interfaces, used on everything from humanoids to embedded devices.

iDynTree

iDynTree is a library of robots dynamics algorithms for control, estimation and simulation.

iDynTree is specifically designed for free-floating robots, but it is possible to use it also with fixed-base robots.

iDynTree is written in C++ language, but thanks to SWIG it is possible to use the iDynTree algorithms in several other languages. Support and documentation is provided in particular for C++, Matlab, Python and Lua.

whole-body-controllers

The codebase whole-body-controllers contains Simulink-based whole-body controllers developed to control the iCub humanoid robot. It can be imagined as a starting point and a support repository for a user that intends to develop a new Simulink controller (not necessarily for the iCub robot) in within the framework of the robotology organisation. It is worth noting that:

• The controllers stored in this repository are an overview of the possibile control frameworks that can be implemented using the robotology software infrastructure. Also, the repository contains a library of configuration and utility Matlab functions to design simulations with Gazebo simulator and on the real robot iCub.

• The robot dynamics and kinematics is computed run-time by means of WBToolbox, a Simulink library that wraps iDyntree. For more information on iDyntree library, see also this README.

• The Simulink models implement different control strategies both for fixed-base and for floating-base robots. They space from momentum-based torque control to inverse-kinematics-based position control. Have a look at the controllers folder for more details.

gazebo-yarp-plugins

The gazebo-yarp-plugins are a set of Gazebo plugins that expose YARP interfaces for robot simulated in Gazebo. The main design objective for the gazebo-yarp-plugins is to provide the exact same interfaces that can be found in real world YARP-powered robot in the Gazebo simulation, to ensure that YARP-based software can run with no modifications both in simulation and on the real robots.