Deepmind announced opening of Mujoco physical processes simulator

Google owned by Deepmind , who gained fame with its developments in the field of artificial intelligence and building neural networks capable of playing computer games At the human level, announced about opening the engine to simulate the physical processes Mujoco (Multi-Joint Dynamics WITH CONTACT). The engine is aimed at modeling the articulated structures interacting with the environment, and is used to simulate when developing robots and artificial intelligence systems, at the stage before the embodiment of the developed technology in the form of a finished device.

The code is written in C / C ++ languages ​​and will published under the APACHE 2.0 license. Linux, Windows and MacOS platforms are supported. Work on the opening of all source-related source projects is planned to be completed in 2022, after which Mujoco will switch to the open development model, implying the possibility of participating in the development of community representatives.

Mujoco is a library with the implementation of the mechanization of the simulation of physical processes of general purpose, which can be applied in the process of research and development of robots, biomechanical devices and machine learning systems, as well as when creating graphics, animations and computer games. The simulation engine is optimized to achieve maximum performance and allows you to manipulate objects at a low level, while providing high accuracy and ample modeling capabilities.

models are determined using the MJCF scene description language based on XML and compiled using a special optimizing compiler. In addition to the MJCF, the engine supports the download of files in Unified Robot Description Format. Mujoco also provides a graphical interface for interactive 3D-visualization of the process of simulation and rendering results using OpenGL.



Main features:

/Media reports.