Robotics & embodied agents
809 agents · most starred first
Robotics & embodied agents, page 6 of 9
Reachy Mini MCP | Give your AI a body. This MCP server lets AI systems control the Pollen Robotics Reachy Mini robot. | Speak, listen, see you, and express emotions through physical movement and voice. | Works with Claude, Windsurf, Cursor, or any MCP-compatible AI. | Zero robotics expertise required.
Solving openAI's game 'BipedalWalker-v2' with Deep Reinforcement Learning
A Reinforcement Learning agent to perform overtaking action using Double DQN based CNNs which takes images as input built using TensorFlow.
Autonomous vehicle training environment with configurable difficulty based on ML-Agents v0.6
Multi-agent-path-planning by Python,with 4 entrances, 4 target and 8 AGVs
[ICDE 2022] Human-Drone Collaborative Spatial Crowdsourcing by Memory-Augmented Distributed Multi-Agent Deep Reinforcement Learning
[ICDE 2023] Exploring both Individuality and Cooperation for Air-Ground Spatial Crowdsourcing by Multi-Agent Deep Reinforcement Learning
This repository contains software for multi-agent simulation model of mixed traffic flow of connected (HVs) and automated vehicles (AVs) in Python using pygame, matplotlib, numpy, scipy and seaborn libraries. The software is capable of simulating many different cases of traffic flow and creates data files and figures for the purpose of analysis.
Implementation code for: O. Natan and J. Miura, “End-to-end Autonomous Driving with Semantic Depth Cloud Mapping and Multi-agent,” IEEE Transactions on Intelligent Vehicles, vol. 8, no. 1, pp. 557-571, Jan. 2023.
Code of the ICCV 2023 paper "March in Chat: Interactive Prompting for Remote Embodied Referring Expression"
Safe Multi-Agent Robosuite benchmark for safe multi-agent reinforcement learning research.
NICS FAST-LIO-SAM branch. With FAST-LIO 2 fount-end and LIO-SAM back-end. Add Multi-agent support and Hesai Lidar support,
UAVs path planning algorithm based on multi-agent deep reinforcement learning algorithm MASAC
A modular simulation framework for controlling multiple drones in a PX4-based environment. It features ROS 2 integration, G-code trajectory planning, multi-agent communication, and a Dockerized setup for easy deployment and visualization.
Risk-aware multi-agent deep reinforcement learning for packet routing in ultra-dense LEO satellite networks
158 expert AI skills for ROS 2, Nav2, perception, SLAM, localization, URDF, Gazebo, and robotics development with GitHub Copilot.
Multi-agent version of highway_env simulator
Implementation of the paper "Reciprocal Velocity Obstacles for real-time multi-agent navigation" in ROS using rospy in Turtlesim
it provides Pepper Robot conversation abilities to handle a free open-domain dialogue.
Distributed (and single-agent) reinforcement learning code for simulated multi-robot distributed assembly.
包括UGV和UAV在内的异构混合阶多智能体系统的一致性
Implement a single- angle solver, namely space-time A*, and parts of three MAPF solvers, namely prioritized planning, Conflict-Based Search (CBS), and CBS with disjoint splitting.
Dijkstra+minimum jerk+ORCA(RVO2)
Development of parametric, deep learning, and reinforcement learning agent-based model of car-following behaviour. The models aim to be data-driven, and take into account the heterogeneity of drivers' behaviour and vehicle types. Potential future applications include investigation of a mixed traffic of human-driven vehicles and autonomous vehicles.
Simulation of a distributed control algorithm to drive a multi-agent system with a single leader (has knowledge about target locations) to the target positions with collision avoidance.
[TPAMI 2026] UniMM: A Unified Mixture Model Framework for Multi-Agent Simulation
ADOS Drone Agent — Software-defined drone network agent. Install on any companion computer to make your drone smart, autonomous, and fleet-ready.
A general-purpose robotic agent framework based on LLMs. The LLM can independently reason, plan, and execute actions to operate diverse robot types across various scenarios to complete unpredictable, complex tasks.
The proposed platform is an AI-native middleware designed to integrate physical infrastructure into the Agentic Economy. This toolkit allows any hardware entity, from robotic arms to autonomous drones and DePIN networks, to act as a financially sovereign and legally compliant agent.
Multi-agent occlusion inference using observed driver behaviors. A driver sensor model is learned using a conditional variational autoencoder. Information from multiple drivers is fused into an ego vehicle's map using evidential theory.
Lightweight and Effective Preference Construction in PIBT for Large-Scale Multi-Agent Pathfinding (SoCS-25)
Codes for the paper "AoI Aware Radio Resource Management of Autonomous Platoons via Multi Agent Reinforcement Learning"
A multi-device environment for evaluating intelligent vehicle interaction across connected cockpit systems
Zephyrine: An augmented Agentic Assistant GNC framework system for experimental Aircraft. Engineered for processing of navigation loops and sensor fusion and adaptation to bridge the gap between strategic virtual path and action planning and real-world kinematic control.
Agents code for Multi-Agent Connected Autonomous Driving (MACAD) described in the paper presented in the Machine Learning for Autonomous Driving Workshop at NeurIPS 2019:
Multi agent system to coordinate multiple UAVs
Official implementation for the LABOR (LAnguage-model-based Bimanual ORchestration) Agent.
Official implementation of LLM+MAP: Bimanual Robot Task Planning using Large Language Models (LLMs) and Planning Domain Definition Language (PDDL).
The system enables sophisticated coordination of multiple drones through natural language commands, visual inputs, and real-time environmental data.
Real-time drone swarm autonomy simulation using Jev for fast System 1 reflex decisions and collision avoidance, with optional System 2 reasoning for strategic guidance
Multi-Robot Exploration, Mapping and Navigation using ros2
Multi-Agent Reinforcement Learning for UAV-USV Cooperative Task Allocation in Offshore Wind Farms
The visual compute platform for AI and robotics — connecting agents, GPU workloads, simulation, training, and real robots through typed workflows.
Offline fault diagnosis assistant for the AgiBot X2 Ultra humanoid: drop in a log package, get a three-round reviewed diagnosis. Rules + causal inference + LLM agent.
RoboClaw live robot lab — community, experience, and API. Star to unlock the Host Concierge.
Agentic AI Robot: Industrial Safety Monitoring and Control
Open-source Android framework for low-latency, LLM-driven multimodal interaction on Pepper. Uses end-to-end speech-to-speech models and extensive Function Calling for agentic robot control (navigation, gaze, vision, touch). Also runs on regular Android devices.
Agentic Robotic Software System
Local agent skill for ROS 2 robot control
Control TurtleBot3 with natural language using LLMs
A source-aware directory of Embodied AI agents, LLM/VLM systems, VLA models, and robotic manipulation projects.
Manipulation benchmarks test acting to change; EmbodiedLab tests acting to know. Given only an objective and the apparatus, a robot runs its own experiment in a physics-based lab to find out a result it cannot know in advance. 35 science experiments, scored exactly.
Deep Reinforcement Learning Toolbox for Robotics using Keras and TensorFlow
Gym env for TurtleBot3 robot
Autonomous robotics research with simulation feedback
Multi-UAV coordination console for manual and LLM-driven missions
(ICRA 2025) Inverse Mixed Strategy Games with Generative Trajectory Models
Repository for "Who Plays First? Optimizing the Order of Play in Stackelberg Games with Many Robots" - RSS 2024
Official implementation of "Learning to Assist: Physics-Grounded Human-Human Control via Multi-Agent Reinforcement Learning" (CVPR 2026)
2025 Student Competition organized by the Department of Artificial Intelligence and System Engineering. I secured 3rd Place out of 18 participants in this competitive algorithmic challenge.
[ICCV'21] Official implementation of the "Social NCE: Contrastive Learning of Socially-aware Motion Representations" in PyTorch, for Social-STGCNN on trajectory forecasting.
BipedalWalker - AI teaching itself to walk
Natural-language pick-and-place: a Claude VLM grounds objects in RGB-D, an 8-tool agent loop plans, MoveIt 2 executes on a simulated UR3e behind a safety gate.
[CVPR2026] InterAgent: Physics-based Multi-agent Command Execution via Diffusion on Interaction Graphs
🦾 Bridge OpenClaw AI agents with physical servo robots — Python library for LOBOT LX bus servo control, motion teaching & playback
OpenAnimal OpenBrain Emotional Robots and Droids - Free Unreal Engine Plugin 🢗
An advanced embodied AI agent with personality-driven interactions, multi-modal perception (vision, speech), autonomous task execution, and real-time voice synthesis. Inspired by robotics subsumption architectures, it bridges cognitive reasoning, behavioral decision-making, and virtual/physical body control for immersive AI experiences.
Embodied-OS:面向具身智能的分层运行时——Driver·Capability·Skill·Brain 全栈参考实现
A flexible development platform for Pioneer Multi-Robot Systems (MRS) http://wiki.ros.org/pioneer_mrs
Python project for the paper "Adversarial Deep Reinforcement Learning for Improving the Robustness of Multi-agent Autonomous Driving Policies".
multi-agent car parking using reinforcement learning
Robot data infrastructure and runtime: turn video, state, actions, and telemetry into training episodes, RL environments, policy rollouts, and replayable evidence.
AV2 Scenario Mining Challenge on CVPR Workshop 2026
Python and MATLAB codes for Density-Driven Optimal Control (D2OC) using Optimal Transport and Wasserstein distance, enabling decentralized multi-agent / multi-robot non-uniform area coverage.
[IROS'25] COCMT
Decentralized multi-agent collision avoidance
[ICML 2024] Official repository of ICML 2024 - RoboMP2: A Robotic Multimodal Perception-Planning Framework with Multimodal Large Language Models
A scheduler to manage a multi tool dual arm robot while avoiding arm-to-arm collisions; considering complex side constraints; and optimizing for cyclic schedules.
Robotic arm controled via OpenClaw. Natural language to motion. Where Openclaw having arm
KECRA (Knowledge-Enhanced Cobot Reasoning Assistant ) is a framework for language-based control of collaborative robots (cobots) in dynamic construction environments. It combines Large Language Models (LLMs) with Visual Foundation Models (VFMs) to allow human operators to guide robotic assistance through natural language dialogues.
AI-powered Mars rover simulation using Hermes Agent (Nous Research)
Annotate any LeRobot dataset with a swarm of agents. Turn raw demos into training-ready data for robotic foundation models.
Natural language to ROS2 actions, a structured LLM grounding engine for any robot.
An AI agent flies ArduPilot over MAVLink. Every decision, command and fault streams to Voight as one trace.
Agent-readable knowledge base for reproducing and operating embodied-AI simulation toolchains. Five-layer docs (quickstart / concepts / field notes / troubleshooting / code), plus a cross-project layer of reproduction workflow, issue taxonomy, selection guide, best practices and terminology mapping.
RoboQuest is a benchmark for goal-directed embodied exploration, in which a mobile manipulator must gather task-relevant information through physical interaction, act on the evidence, and decide for itself when the task is done.
Next-Generation Gyms for Embodied AI Agents, on Bittensor
Open-source safety and cognitive control infrastructure for physical AI agents — robots, LLM agents, autonomous systems.
Conflict Avoidant CBS (CA-CBS)
LLM-IK: Solving Inverse Kinematics using Large Language Models
Companion repo for the survey "A Survey of Agentic Robotics: Toward Continual Self-Improvement": a taxonomy and paper list of foundation-model agents that orchestrate robot tasks, generate motion, and improve robot systems.
Real-time quadruped robot gait telemetry analyzer & stability evaluator (Unitree / Boston Dynamics)
🛞 scout; a Strands agent that sees, thinks, talks & drives a FrodoBots Earth Rover Mini+. Vision · language · action. ECoT dataset on 🤗 cagataydev/scout-earthrover-ecot
Official Implementation of "Systematic Multi-Agent Vision-and-Language Navigation: Formulation, Benchmark, and Method"
A robot agent harness for visual action selection, subagent context management, and learning from execution through persistent playbooks.
GeoTask — 面向 AI Agent 的时空任务上下文引擎 | Spatiotemporal Task Context Engine for AI agents.
Natural-language PTZ camera agent that runs at the edge, with a 536-task Blender digital-twin benchmark (HRI '26).
Eatable: From a meal... to a life of independence.
A ROS2 node that uses OpenAI GPT Assistant to generate navigation instructions for nav2 🐁🤖🇮🇹🌍🏆