Robotics & embodied agents
Robotics and embodied agents control physical hardware or simulate it: quadcopters, humanoids, quadrupeds, and drones. Some, like dimos, translate natural language commands into robot actions, while others such as habitat-lab provide simulated environments for training embodied agents on navigation and manipulation tasks.
809 agents · most starred first
Top 100 of 809
A customisable 3D platform for agent-based AI research
Dimensional is the agentic operating system for physical space. Command humanoids, quadrupeds, drones, and other hardware platforms in natural language and build multi-agent systems that work seamlessly with physical input (cameras, lidar, actuators).
A modular high-level library to train embodied AI agents across a variety of tasks and environments.
🐕 MindPaw — 基于 ESP8266 的开源桌面四足机器狗,支持语音、手势、网页遥控、豆包 AI 对话与 PAD 情感交互。可选 AI Infra 网关,适合机器人入门、嵌入式 AI、边缘 Agent 与 HRI 研究。
An efficient single/multi-agent trajectory planner for multicopters.
PyBullet Gymnasium environments for single and multi-agent reinforcement learning of quadcopter control
This is a curated list of "Embodied AI or robot with Large Language Models" research. Watch this repository for the latest updates! 🔥
PhyAgentOS is a self-evolving embodied AI operating system built on agentic workflows.
RPent: Agentic Infrastructure for the Physical World
PhyAgentOS is a self-evolving embodied AI operating system built on agentic workflows.
PhyAgentOS is a self-evolving embodied AI operating system built on agentic workflows.
PhyAgentOS is a self-evolving embodied AI operating system built on agentic workflows.
ROSA 🤖 is an AI Agent designed to interact with ROS1- and ROS2-based robotics systems using natural language queries. ROSA helps robot developers inspect, diagnose, understand, and operate robots.
Python implementation of a bunch of multi-robot path-planning algorithms.
A Python-based lightweight robot simulator designed for navigation, control, and learning
A generative and self-guided robotic agent that endlessly propose and master new skills.
Scalable Multi-Agent RL Training School for Autonomous Driving
OpenAI Gym environments for an open-source quadruped robot (SpotMicro)
Library with search algorithms for task and path planning for multi robot/agent systems
A Multi-Agent Reinforcement Learning Environment for Large Scale City Traffic Scenario
[ICML 2026] A Framework for Benchmarking and Improving Coding Agents for Robot Manipulation
A unified framework for building, evaluating, and recursively improving embodied agents. 🚀 Code release coming soon.
SimWorld: An Open-ended Realistic Simulator for Autonomous Agents in Physical and Social Worlds
[CVPR 2022] HiVT: Hierarchical Vector Transformer for Multi-Agent Motion Prediction
An open harness to simulate and deploy PX4/ArduPilot/Betaflight perception-based drone swarms with ROS2, YOLO, NVIDIA Jetson
Multi-Agent Transport Simulation
API to run VirtualHome, a Multi-Agent Household Simulator
This repository is a paper digest of recent advances in collaborative / cooperative / multi-agent perception for V2I / V2V / V2X autonomous driving scenario.
VMAS is a vectorized differentiable simulator designed for efficient Multi-Agent Reinforcement Learning benchmarking. It is comprised of a vectorized 2D physics engine written in PyTorch and a set of challenging multi-robot scenarios. Additional scenarios can be implemented through a simple and modular interface.
Trajectory Planner in Multi-Agent and Dynamic Environments
An open agentic toolkit for physics-based Real2Sim of Robot Manipulation Tasks
1 million FPS multi-agent driving simulator
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.
Implementation of Gamma-World: Generative Multi-Agent World Modeling Beyond Two Players
The open-source operating system for physical AI.
Deep Reinforcement Learning (PPO) in Autonomous Driving (Carla) [from scratch]
A robotic harness that gives LLM agents a visual-geometric control panel so that they can directly understand and invoke embodied tasks.
A framework where a deep Q-Learning Reinforcement Learning agent tries to choose the correct traffic light phase at an intersection to maximize traffic efficiency.
[ICML 2024] LEO: An Embodied Generalist Agent in 3D World
A large-scale benchmark for evaluating vision-language-action models, embodied agents, and vision-language models
Just a VLM Agent Can Play Robots
COVINS-(G) -- A (Generic) Framework for Collaborative Visual-Inertial SLAM and Multi-Agent 3D Mapping
[NeurIPS 2026] SpatialClaw: Rethinking Action Interface for Agentic Spatial Reasoning
PRIMAL: Pathfinding via Reinforcement and Imitation Multi-Agent Learning -- Distributed RL/IL code for Multi-Agent Path Finding (MAPF)
Anonymous Multi-Agent Path Finding (MAPF) with Conflict-Based Search and Space-Time A*
Multi-Robot Warehouse (RWARE): A multi-agent reinforcement learning environment
A universal flight control tuning framework
Gym Electric Motor (GEM): An OpenAI Gym Environment for Electric Motors
[CVPR2024 Highlight] Editable Scene Simulation for Autonomous Driving via LLM-Agent Collaboration
MIGHTY: Hermite spline-based real-time trajectory planner for single- and multi-agent UAVs and ground robots (ROS 2). Accepted to IEEE RA-L.
multi-agent deep reinforcement learning for large-scale traffic signal control.
A unified, agentic system for general-purpose robots, enabling multi-modal perception, mapping and localization, and autonomous mobility and manipulation, with intelligent interaction with users.
An Efficient Multi-Agent Path Finding Solver for Car-Like Robots
The Agentic Operating System for Robots
JARVIS-1: Open-world Multi-task Agents with Memory-Augmented Multimodal Language Models
面向硬件产品PCB方案设计的AI Agent,Agent会自动帮你进行需求确认,实时分析国内外各类芯片技术方案,进行器件选型,下载datasheet,输出BOM表,计算价格,输出模块原理图,最终整合成可落地技术方案。
EMERGE-Policy: A Robot Mind Emerges Beyond a Single Policy
Benchmark for Continuous Multi-Agent Robotic Control, based on OpenAI's Mujoco Gym environments.
Multi-Agent Connected Autonomous Driving (MACAD) Gym environments for Deep RL. Code for the paper presented in the Machine Learning for Autonomous Driving Workshop at NeurIPS 2019:
ENPIRE is an agentic framework that allows AI coding agents to autonomously self-improve real-world robot policies through closed-loop physical trial, error, and evolution
[AAAI 2026 Oral] Cook and Clean Together: Teaching Embodied Agents for Parallel Task Execution
GPT-Policy: In-Context Robot Learning with VLM Agents
The open-source operating system for robots — install it and your robot comes alive
[CVPR 2025]MAGiC-SLAM: Multi-Agent Gaussian Globally Consistent SLAM
Open Platform for Embodied Agents
Open-source MAVLink fleet operations for PX4: SITL, drone shows, search and rescue, cooperative autonomy, and field validation
A C++/Python implementation of the StreetLearn environment based on images from Street View, as well as a TensorFlow implementation of goal-driven navigation agents solving the task published in “Learning to Navigate in Cities Without a Map”, NeurIPS 2018
[ICCV 2021] Official PyTorch Implementation of "AgentFormer: Agent-Aware Transformers for Socio-Temporal Multi-Agent Forecasting".
Train embodied agents that can answer questions in environments
SIMPL: A Simple and Efficient Multi-agent Motion Prediction Baseline for Autonomous Driving
A set of high-dimensional continuous control environments for use with Unity ML-Agents Toolkit.
Open-Source Framework for Development, Simulation and Benchmarking of Behavior Planning Algorithms for Autonomous Driving
Embodied Co-Design for Rapidly Evolving Agents: Taxonomy, Frontiers, and Challenges
[ICLR-2025] POGEMA stands for Partially-Observable Grid Environment for Multiple Agents. This is a grid-based environment that was specifically designed to be flexible, tunable and scalable. It can be tailored to a variety of PO-MAPF / MAPF settings.
The first exam for code agents in the physical world: frozen robot tasks from many benchmark suites, each answered with a program and verified in a separate sandbox. Runs on Harbor.
A Language Agent for Autonomous Driving
Embodied Agent Interface (EAI): Benchmarking LLMs for Embodied Decision Making (NeurIPS D&B 2024 Oral)
[ICLR 2024] Source codes for the paper "Building Cooperative Embodied Agents Modularly with Large Language Models"
On the Road with GPT-4V(ision): Explorations of Utilizing Visual-Language Model as Autonomous Driving Agent
[NeurIPS 2024] SMART: Scalable Multi-agent Real-time Motion Generation via Next-token Prediction
A data-driven, fast driving simulator for multi-agent coordination under partial observability.
[RA-Letter 2022] Reinforcement Learned Distributed Multi-Robot Navigation with Reciprocal Velocity Obstacle Shaped Rewards
The communication layer between people and robots, and between robots. Talk to any robot in your own words, with an LLM for its brain. Robots that have never met check each other's passports and share the work. Microduck, Huebot, Open Duck Mini, LeRobot arms, XLeRobot carts, AlohaMini, ToddlerBot, MOSS, ROS bases and industrial arms on ROS 2.
Pytorch implementation of CVPR2020 paper “VectorNet: Encoding HD Maps and Agent Dynamics from Vectorized Representation”
Official code for article "LLMLight: Large Language Models as Traffic Signal Control Agents".
Survey on Multimodal Embodied Agents: Computer-Use → Robot-Use
Write your robot software in minutes.
DEPRICATED! Use the multi-agent suite instead!
Spectrum sharing in vehicular networks based on multi-agent reinforcement learning, IEEE Journal on Selected Areas in Communications
Continuous CBS - a modification of conflict based search algorithm, that allows to perform actions (move, wait) of arbitrary duration. Timeline is not discretized, i.e. is continuous.
Browser teleoperation data for embodied AI — elizaOS-ready episodes, trajectory_db converter, onchain provenance. The fuel station for agent robot stacks.
llama.cpp (GGUF LLMs) and llava.cpp (GGUF VLMs) for ROS 2
Fewer Tokens, Better Action: GPT-6 Astra Robot Agents with 14% Higher Success but 65% Fewer Tokens
Codebase for paper: RoCo: Dialectic Multi-Robot Collaboration with Large Language Models
Use Reinforcement Learning to train an autonomous driving agent in CARLA Simulator
This is a multi-agent path planning(also known as Multi-Agent Path Finding, MAPF) algorithm package for ROS/ROS2.
Framework for developing OpenAI Gym robotics environments simulated with Ignition Gazebo
Agent skill for production-grade ROS 2 development. Progressive-disclosure SKILL.md covering workspace, nodes, executors, QoS, ros2_control, Nav2, MoveIt 2, real-time, and deployment. Works with Claude Code, Codex, Cursor, Gemini CLI.
76 audit-ready Claude skills automating the industrial robot, cobot, AMR, ROS2, V&V, AI/ML, and IEC 62443 lifecycle. 38 builder + reviewer pairs anchored to ISO 10218, 13849, 62061, 12100, 9283, 15066, 3691-4, and IEC 62443.
a comprehensive and critical synthesis of the emerging role of GenAI across the full autonomous driving stack