embodied-agents

RAI is a vendor agnostic agentic framework for Physical AI robotics, utilizing ROS 2 tools to perform complex actions, defined scenarios, free interface execution, log summaries, voice interaction and more.

Seamlessly integrate state-of-the-art transformer models into robotics stacks

An embodied task agent that connects perception, action, verification, and learning in a continuous physical-world loop

Safety in Embodied AI: A Survey of Risks, Attacks, and Defenses | 500+ Papers | Perception, Cognition, Planning, Interaction, Agentic System

A curated list of Embodied Recursive Self-Improvement (Embodied RSI) research, systems, benchmarks, and industry updates.

EMOS CLI Utility

VLM-driven 3D digital-twin embodied simulation. A minimal, locally runnable framework for research and education (SD-AgentFoundry-3D). VLM駆動エージェントによる3次元デジタルツイン身体化シミュレーション。ローカル実行・最小構成の研究/教育用基盤（SD-AgentFoundry-3D）

AEROS: A Runtime Architecture for Embodied Agents with Installable Capability Modules

Simulation code for: Federated Single-Agent Robotics — Fleet Coordination Under Governed Autonomy

Simulation code for: Harnessing Embodied Agents: Runtime Governance for Policy-Constrained Execution

Controlled evaluation of persistent object memory for partially observable embodied agents in AI2-THOR.

Compare physical and virtual embodiment in negotiation: fruit-sharing profiles, paired protocols and analysis examples. IVA 2023.

Experiment code for: Learning Without Losing Identity — Capability Evolution for Embodied Agents

Simulation code for: Governed Capability Evolution for Embodied Agents

Recursive self-improvement for hybrid agents: digital and physical experience improving each other, with an evidence ledger that refuses unverified progress

Experiment code for: ECM Contracts — Contract-Aware, Versioned, and Governable Capability Interfaces for Embodied Agents