SimBenchmark

Created in June 08, 2018 by Dongho Kang

2018   ·   robotics   physics simulation   ·   software


SimBenchmark is a benchmark suite for state-of-the-art physics engines. We assessed several widely used physics engines for robotics and machine learning applications:

  • RAISIM
  • Bullet Physics
  • Open Dynamics Engine
  • Multi-Joint dynamics with Contact (a.k.a. MuJoCo)
  • DART Sim

More details are available at the SimBenchmark website



Enjoy Reading This Article?

Here are some more articles you might like to read next:

  • Learning Steerable Imitation Controllers From Unstructured Animal Motions
  • Tuning Legged Locomotion Controllers via Safe Bayesian Optimization
  • RL + Model-based Control: Using On-demand Optimal Control to Learn Versatile Legged Locomotion
  • Animal Motions on Legged Robots Using Nonlinear Model Predictive Control
  • Nonlinear Model Predictive Control for Quadrupedal Locomotion Using Second-Order Sensitivity Analysis