robosuite: A Modular Simulation Framework and Benchmark for Robot Learning

Abstract

robosuite is a simulation framework for robot learning powered by the MuJoCo physics engine. It offers a modular design for creating robotic tasks as well as a suite of benchmark environments for reproducible research. This paper discusses the key system modules and the benchmark environments of our new release robosuite v1.5.

For more information, please visit https://robosuite.ai

References in corpus (3)

• Soft Actor-Critic Algorithms and Applications
• Deep Reinforcement Learning that Matters
• AI2-THOR: An Interactive 3D Environment for Visual AI

Cited by in corpus (36)

• Orbit: A Unified Simulation Framework for Interactive Robot Learning Environments
• A Survey of Zero-shot Generalisation in Deep Reinforcement Learning
• What Matters in Learning from Offline Human Demonstrations for Robot Manipulation
• iGibson 2.0: Object-Centric Simulation for Robot Learning of Everyday Household Tasks
• Residual Learning from Demonstration: Adapting DMPs for Contact-rich Manipulation
• iGibson 1.0: a Simulation Environment for Interactive Tasks in Large Realistic Scenes
• BEHAVIOR: Benchmark for Everyday Household Activities in Virtual, Interactive, and Ecological Environments
• Conservative Safety Critics for Exploration
• Benchmarking the Sim-to-Real Gap in Cloth Manipulation
• ManiSkill: Generalizable Manipulation Skill Benchmark with Large-Scale Demonstrations
• Long-Horizon Visual Planning with Goal-Conditioned Hierarchical Predictors
• A Survey of Embodied AI: From Simulators to Research Tasks
• Knowledge Transfer for Cross-Domain Reinforcement Learning: A Systematic Review
• On the Role of the Action Space in Robot Manipulation Learning and Sim-to-Real Transfer
• ThriftyDAgger: Budget-Aware Novelty and Risk Gating for Interactive Imitation Learning
• Geometric Impedance Control on SE(3) for Robotic Manipulators
• SECANT: Self-Expert Cloning for Zero-Shot Generalization of Visual Policies
• 3DB: A Framework for Debugging Computer Vision Models
• Contact-rich SE(3)-Equivariant Robot Manipulation Task Learning via Geometric Impedance Control
• Human-in-the-Loop Imitation Learning using Remote Teleoperation
• CORA: Benchmarks, Baselines, and Metrics as a Platform for Continual Reinforcement Learning Agents
• Lifelong Robotic Reinforcement Learning by Retaining Experiences
• rl_reach: Reproducible Reinforcement Learning Experiments for Robotic Reaching Tasks
• Continual Model-Based Reinforcement Learning with Hypernetworks
• Zero-shot Task Adaptation using Natural Language
• S4RL: Surprisingly Simple Self-Supervision for Offline Reinforcement Learning
• Leveraging Demonstrator-perceived Precision for Safe Interactive Imitation Learning of Clearance-limited Tasks
• Guided Imitation of Task and Motion Planning
• Multi-Modal Mutual Information (MuMMI) Training for Robust Self-Supervised Deep Reinforcement Learning
• Building Minimal and Reusable Causal State Abstractions for Reinforcement Learning
• Koopman Q-learning: Offline Reinforcement Learning via Symmetries of Dynamics
• Learning Robotic Manipulation Skills Using an Adaptive Force-Impedance Action Space
• Learning Time-Invariant Reward Functions through Model-Based Inverse Reinforcement Learning
• Learning a Skill-sequence-dependent Policy for Long-horizon Manipulation Tasks
• Robust Robotic Control from Pixels using Contrastive Recurrent State-Space Models
• Bottom-Up Skill Discovery from Unsegmented Demonstrations for Long-Horizon Robot Manipulation