Skip to main content

The Cognitive BIOS for Robotics - Standard CLI

Project description

OpenRGD: A Proposed Standard for Cognitive Embodiment

Status: RFC License: MIT

OpenRGD (Robot Graph Definition) is an open architectural standard designed to serve as the universal semantic bridge between Artificial Intelligence and Physical Reality.
It defines a machine‑readable “Self‑Model” that enables any cognitive agent to safely and effectively embody any physical machine.


⚡ Start Instantly: The 30‑Second Magic

Don't just read the spec — experience the Cognitive BIOS immediately.

🐧 Linux / macOS

# 1. Install the Toolchain
git clone https://github.com/OpenRGD/openrgd.git && cd openrgd && pip install -e .

# 2. The Magic Sequence (Create → Enter → Boot)
rgd init Robot && cd Robot && rgd boot

🪟 Windows (PowerShell)

# 1. Install the Toolchain
git clone https://github.com/OpenRGD/openrgd.git ; cd openrgd ; pip install -e .

# 2. The Magic Sequence (Create → Enter → Boot)
rgd init Robot ; cd Robot ; rgd boot

1. Interoperability

A unified language enabling cognitive models to transfer seamlessly across different robotic bodies, platforms, and ecosystems.


2. The Architecture

OpenRGD is not a monolithic file but a hierarchical Semantic Graph structured into six normative domains.

01. Foundation — The Body

Immutable physical reality.
Defines hardware abstraction: actuator models, sensors, dimensions, topology.

02. Operation — The Physiology

Autonomic safety system.
Defines operational envelopes, reflex rules, and constraints that override unsafe intent.

03. Agency — The Mind

Capability interface.
Defines the “Skills Library” and how the cognitive agent interacts with the environment.

04. Volition — The Conscience

Intent alignment.
Defines ethics, value hierarchies, and decision‑governance logic.

05. Evolution — The Lifecycle

Temporal state.
Tracks wear, fatigue, maintenance, lifespan, and change over time.

06. Ether — The Society

Collective intelligence.
Defines swarm behavior, consensus, inter‑agent communication, and reputation.


3. Toolchain (CLI)

The rgd CLI manages the entire lifecycle of an embodiment definition — effectively acting as the Operating System of the Standard.

The Golden Loop

  • Ingest (rgd import) — Convert URDF/USD into OpenRGD.
  • Compile (rgd compile-spec) — Produce Digital Twins (.jsonc for LLMs, .json for validation).
  • Bridge (rgd export) — Export to ROS2, Isaac Lab, and other engines.
  • Boot (rgd boot) — Initialize the Cognitive BIOS and generate the System Prompt.

Example Workflow

rgd import my_robot.urdf
cd my_robot
rgd compile-spec
rgd export ros2
rgd boot

4. The Kernel Concept

Implementations rely on a Semantic Kernel, an orchestrator that loads the six domains and guarantees a complete and consistent cognitive embodiment before any action is executed.


5. Governance

OpenRGD follows an open RFC‑driven development model.

  • Status: Draft v0.1
  • Documentation: See STRUCTURE.md
  • Contribution Guide: See CONTRIBUTING.md

6. Future Roadmap: AI‑Governed Evolution

This section anticipates the v1.0 architecture.

OpenRGD is designed to support AI‑assisted stewardship via Commitment Proposals (CPs) — machine‑generated diffs that undergo:

  1. Automated Validation
  2. Human Maintainer Review
  3. Regulatory / Cryptographic Audit

A controlled path toward shared evolution of safety standards.


7. Authors

  • Pasquale Ranieri (Italia Robotica) - Lead Architect & Specification Author

Acknowledgements

  • Recursive Design: Portions of this specification were developed with the assistance of Large Language Models (LLMs). The latest generation of models shows an emerging ability to infer physical behavior—linking geometry, dynamics, and constraints in ways that approximate real-world reasoning. These capabilities helped validate the internal consistency of the 01_Foundation domain.

8. Contribute

OpenRGD is an open standard. To participate in its development, read the CONTRIBUTING.md guide and follow the contribution workflow.


Copyright © 2025 OpenRGD Organization. Distributed under the MIT License.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

rgd-0.1.1.tar.gz (286.6 kB view details)

Uploaded Source

Built Distribution

If you're not sure about the file name format, learn more about wheel file names.

rgd-0.1.1-py3-none-any.whl (345.0 kB view details)

Uploaded Python 3

File details

Details for the file rgd-0.1.1.tar.gz.

File metadata

  • Download URL: rgd-0.1.1.tar.gz
  • Upload date:
  • Size: 286.6 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.12.10

File hashes

Hashes for rgd-0.1.1.tar.gz
Algorithm Hash digest
SHA256 5c4f19abc31bc2886c8c0318e8f01a0ed84a9c68987b4f71a3e37837b6e20a31
MD5 e560eaa7b5c289de379466b6ece58e6f
BLAKE2b-256 a287dcf67c3534b38dd7d08ef214723d14c8c0216aa5772630e563cd317e344c

See more details on using hashes here.

File details

Details for the file rgd-0.1.1-py3-none-any.whl.

File metadata

  • Download URL: rgd-0.1.1-py3-none-any.whl
  • Upload date:
  • Size: 345.0 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: twine/6.2.0 CPython/3.12.10

File hashes

Hashes for rgd-0.1.1-py3-none-any.whl
Algorithm Hash digest
SHA256 998278d790c9a69676a8564ffa1b65bef2fd6b66d2598ee688ba4241a8809688
MD5 1649a493ffccd03c555cf8784ed48c51
BLAKE2b-256 0deac0e34c1b27680884a6b716b5f76e7684a634f871688b8e33bff929881abe

See more details on using hashes here.

Supported by

AWS Cloud computing and Security Sponsor Datadog Monitoring Depot Continuous Integration Fastly CDN Google Download Analytics Pingdom Monitoring Sentry Error logging StatusPage Status page