RoboManipBaselines: A Unified Framework for Imitation Learning in Robotic Manipulation across Real and Simulated Environments

2025/09/21 by Murooka, Masaki, Motoda, Tomohiro, Nakajo, Ryoichi +4 · 1 citation
#FOS: Computer and information sciences #Robotics (cs.RO)

paper · doi:10.48550/arxiv.2509.17057

Abstract

RoboManipBaselines is an open framework for robot imitation learning that unifies data collection, training, and evaluation across simulation and real robots. We introduce it as a platform enabling systematic benchmarking of diverse tasks, robots, and multimodal policies with emphasis on integration, generality, extensibility, and reproducibility.

Citations

• Self-Augmented Robot Trajectory: Efficient Imitation Learning via Safe Self-augmentation with Demonstrator-annotated Precision
• ManiFlow: A General Robot Manipulation Policy via Consistency Flow Training
• The Sound of Simulation: Learning Multimodal Sim-to-Real Robot Policies with Generative Audio
• RoboArena: Distributed Real-World Evaluation of Generalist Robot Policies
• TACT: Humanoid Whole-body Contact Manipulation through Deep Imitation Learning with Tactile Modality
• GR00T N1: An Open Foundation Model for Generalist Humanoid Robots
• Eagle 2: Building Post-Training Data Strategies from Scratch for Frontier Vision-Language Models
• FlowPolicy: Enabling Fast and Robust 3D Flow-based Policy via Consistency Flow Matching for Robot Manipulation
• 3D-ViTac: Learning Fine-Grained Manipulation with Visuo-Tactile Sensing
• Sparsh: Self-supervised touch representations for vision-based tactile sensing
• π₀: A Vision-Language-Action Flow Model for General Robot Control
• FF-SRL: High Performance GPU-Based Surgical Simulation For Robot Learning
• ManiSkill3: GPU Parallelized Robotics Simulation and Rendering for Generalizable Embodied AI
• UniT: Data Efficient Tactile Representation with Generalization to Unseen Objects
• PaliGemma: A versatile 3B VLM for transfer
• RoboCasa: Large-Scale Simulation of Everyday Tasks for Generalist Robots
• DROID: A Large-Scale In-The-Wild Robot Manipulation Dataset
• 3D Diffusion Policy: Generalizable Visuomotor Policy Learning via Simple 3D Representations
• Towards Diverse Behaviors: A Benchmark for Imitation Learning with Human Demonstrations
• Universal Manipulation Interface: In-The-Wild Robot Teaching Without In-The-Wild Robots
• THE COLOSSEUM: A Benchmark for Evaluating Generalization for Robotic Manipulation
• RoboHive: A Unified Framework for Robot Learning
• GELLO: A General, Low-Cost, and Intuitive Teleoperation Framework for Robot Manipulators
• RoboAgent: Generalization and Efficiency in Robot Manipulation via Semantic Augmentations and Action Chunking
• LIBERO: Benchmarking Knowledge Transfer for Lifelong Robot Learning
• Learning Fine-Grained Bimanual Manipulation with Low-Cost Hardware
• Diffusion Policy: Visuomotor Policy Learning via Action Diffusion
• Diffusion policy: Visuomotor policy learning via action diffusion
• Contact-Rich Manipulation of a Flexible Object based on Deep Predictive Learning using Vision and Tactility
• Error-Aware Imitation Learning from Teleoperation Data for Mobile Manipulation
• RLDS: an Ecosystem to Generate, Share and Use Datasets in Reinforcement Learning
• Isaac Gym: High Performance GPU-Based Physics Simulation For Robot Learning
• What Matters in Learning from Offline Human Demonstrations for Robot Manipulation
• RLBench: The Robot Learning Benchmark & Learning Environment
• OpenAI Gym
• A survey of robot learning from demonstration

Cited by

• TWINS: A Tactile Wearable Isomorphic Arm Networked System for Contact-Rich Manipulation Learning

Related

• RoboVerse: Towards a Unified Platform, Dataset and Benchmark for Scalable and Generalizable Robot Learning
• RoboMIND: Benchmark on Multi-embodiment Intelligence Normative Data for Robot Manipulation
• Generalized Robot Learning Framework
• RoboCLIP: One Demonstration is Enough to Learn Robot Policies
• Robotic Manipulation via Imitation Learning: Taxonomy, Evolution, Benchmark, and Challenges