Dynamics Randomization Revisited:A Case Study for Quadrupedal Locomotion

by   Zhaoming Xie, et al.

Understanding the gap between simulation andreality is critical for reinforcement learning with legged robots,which are largely trained in simulation. However, recent workhas resulted in sometimes conflicting conclusions with regardto which factors are important for success, including therole of dynamics randomization. In this paper, we aim toprovide clarity and understanding on the role of dynamicsrandomization in learning robust locomotion policies for theLaikago quadruped robot. Surprisingly, in contrast to priorwork with the same robot model, we find that direct sim-to-real transfer is possible without dynamics randomizationor on-robot adaptation schemes. We conduct extensive abla-tion studies in a sim-to-sim setting to understand the keyissues underlying successful policy transfer, including otherdesign decisions that can impact policy robustness. We furtherground our conclusions via sim-to-real experiments with variousgaits, speeds, and stepping frequencies. Additional Details: https://www.pair.toronto.edu/understanding-dr/.


page 1

page 2

page 4


Learning Domain Randomization Distributions for Transfer of Locomotion Policies

Domain randomization (DR) is a successful technique for learning robust ...

Learning and Deploying Robust Locomotion Policies with Minimal Dynamics Randomization

Training deep reinforcement learning (DRL) locomotion policies often req...

Dynamics and Domain Randomized Gait Modulation with Bezier Curves for Sim-to-Real Legged Locomotion

We present a sim-to-real framework that uses dynamics and domain randomi...

Sim2Real Transfer for Reinforcement Learning without Dynamics Randomization

In this work we show how to use the Operational Space Control framework ...

General Robot Dynamics Learning and Gen2Real

Acquiring dynamics is an essential topic in robot learning, but up-to-da...

Reinforcement Learning with Adaptive Curriculum Dynamics Randomization for Fault-Tolerant Robot Control

This study is aimed at addressing the problem of fault tolerance of quad...

Learning Memory-Based Control for Human-Scale Bipedal Locomotion

Controlling a non-statically stable biped is a difficult problem largely...