Linux News

DeepMind Announces MuJoCo Physical Process Simulator

Pinterest LinkedIn Tumblr

Google-owned DeepMind , a company known for its developments in the field of artificial intelligence and the construction of neural networks capable of playing computer games at the human level, announced the discovery of the engine for simulating physical processes MuJoCo (Multi-Joint Dynamics with Contact). The engine is aimed at modeling articulated structures interacting with the environment, and is used for simulation in the development of robots and artificial intelligence systems, at a stage before the implementation of the developed technology in the form of a finished device.

The code is written in C / C ++ and will be published under the Apache 2.0 license. Linux, Windows and macOS platforms are supported. The work on the opening of all the source codes associated with the project is planned to be completed in 2022, after which MuJoCo will switch to an open development model, which implies the possibility of participation in the development of community representatives.

MuJoCo is a library with a general purpose physics simulation engine that can be used in the research and development of robots, biomechanical devices and machine learning systems, as well as in the creation of graphics, animation and computer games. The simulation engine is optimized for maximum performance and allows manipulation of objects at a low level, while providing high accuracy and rich simulation capabilities.

Mattermost 6.0 messaging system Released

Models are defined using the XML-based MJCF scene description language compiled with a dedicated optimizing compiler. In addition to MJCF, the engine supports uploading files in the Unified Robot Description Format (URDF). MuJoCo also provides a graphical interface for interactive 3D visualization of the simulation process and rendering of results using OpenGL.

Key features:

  • Simulation in generalized coordinates , eliminating dislocation of joints.
  • Reverse dynamics , detectable even when there is contact.
  • Using Convex Programming for Unified Formulation of Constraints in Continuous Time.
  • Ability to set various constraints, including soft touch and dry friction.
  • Simulation of particle systems, fabrics, ropes and soft objects.
  • Actuators (actuators), including motors, cylinders, muscles, tendons, and crank mechanisms.
  • Solver programs based on Newton , conjugate gradient and Gauss-Seidel methods .
  • Possibility of using pyramidal or elliptical friction cones.
  • Use of a choice of methods of numerical integration of Euler or Runge-Kutta.
  • Multithreaded discretization and approximation by the method of finite differences.

Write A Comment