dm_control: Software and Tasks for Continuous Control

2020/06/30 by Yuval Tassa, Saran Tunyasuvunakool, Alistair Muldal +8 · 7 citations
Computer Science · #cs.RO #cs.AI #cs.LG

paper · pdf · doi:10.1016/j.simpa.2020.100022

arXiv admin note: text overlap with arXiv:1801.00690

arxiv created 2020/09/07 · arxiv updated 2020/09/08

Abstract

The dm_control software package is a collection of Python libraries and task suites for reinforcement learning agents in an articulated-body simulation. A MuJoCo wrapper provides convenient bindings to functions and data structures. The PyMJCF and Composer libraries enable procedural model manipulation and task authoring. The Control Suite is a fixed set of tasks with standardised structure, intended to serve as performance benchmarks. The Locomotion framework provides high-level abstractions and examples of locomotion tasks. A set of configurable manipulation tasks with a robot arm and snap-together bricks is also included. dm_control is publicly available at https://www.github.com/deepmind/dm_control

Citations

• DeepMind Control Suite

Cited by

• Towards Batch-to-Streaming Deep Reinforcement Learning for Continuous Control
• PhyLatent: Learning Dynamics-Relevant Representations for JEPA World Models
• Dream-MPC: Gradient-Based Model Predictive Control with Latent Imagination
• Zero-Shot Whole-Body Humanoid Control via Behavioral Foundation Models
• FLoRA: Sample-Efficient Preference-based RL via Low-Rank Style Adaptation of Reward Functions
• Neural Motion Simulator: Pushing the Limit of World Models in Reinforcement Learning
• Agent-Arena: A General Framework for Evaluating Control Algorithms