This repository serves as a curated list of research papers for the survey "Toward General-Purpose Robots via Foundation Models: A Survey and Meta-Analysis." It targets researchers and engineers in robotics and AI interested in the application of foundation models (FMs) for creating more versatile and capable robots. The primary benefit is a structured overview of the rapidly evolving field, enabling quick identification of relevant work.

How It Works

The repository categorizes papers into two main groups: "Foundation models used in Robotics" and "Robotic Foundation Models." The former includes works that adapt existing FMs (like LLMs, VLMs) for robotic tasks (perception, planning, action). The latter focuses on novel FMs designed specifically for robotics, including generalist agents like GATO and PaLM-E. Papers are included if they feature experiments on real robots, high-fidelity simulations, or real robot datasets.

Highlighted Details

• Comprehensive categorization of papers applying FMs to robotics.
• Includes works on perception, task planning, action generation, data generation, and general-purpose robotic foundation models.
• Papers are dated based on their first arXiv release.
• Focuses on research with experimental validation on physical robots or simulations.

Maintenance & Community

This repository is associated with a survey paper, indicating a structured research effort. The BibTex citation is provided for referencing the survey. Further community engagement or update frequency is not detailed in the README.

Licensing & Compatibility

The repository itself, containing a list of papers, does not appear to have a specific license mentioned. The underlying research papers will have their own respective licenses and terms of use.

Limitations & Caveats

This repository is a static list of papers and does not provide code, models, or direct access to the surveyed research. The "constantly updated" claim is not substantiated with a clear update schedule or mechanism.