A full-stack development platform for AIoT and HMI products that integrates hardware abstraction, multi-runtime app support, and AI agent capabilities for ESP devices.

The ESPHome source code for the Home Assistant Voice: Preview Edition, providing the firmware needed for voice-controlled home automation hardware.

A conversational application for the Reachy Mini robot that integrates real-time voice, vision, and choreographed motion via LLM-powered tool calling.

A physics-rich simulation environment for robotic vision and interaction, featuring a large-scale dataset of articulated objects for part-level understanding.

A unified simulation infrastructure that integrates CARLA and AirSim into a single process to enable high-fidelity, synchronized air-ground embodied intelligence research.

A digital data-generation pipeline that synthesizes 4D human-object interaction trajectories from 3D assets and video priors to train humanoid robots in loco-manipulation.

HumanEgo is a framework for zero-shot robot learning that allows robots to acquire new skills by training on just a few minutes of human egocentric video demonstrations.

An object-centric reinforcement learning policy and benchmark (DexToolBench) for zero-shot dexterous tool manipulation, enabling robots to use various tools in the real world.

An open-source robotics manipulation platform for ROS that provides tools for motion planning, prototyping, and benchmarking algorithms.

Xiaomi-Robotics-1 is a Vision-Language-Action (VLA) foundation model trained on 100K+ hours of trajectories to enable general-purpose mobile manipulation and efficient task adaptation.

An open-source robotics simulator providing high-fidelity physics, rendering, and sensor models for testing and developing robotic systems.

RTAB-Map is a SLAM (Simultaneous Localization and Mapping) library and standalone application that enables robots to map unknown environments and track their position.

SIMPLE is a high‑fidelity simulation environment for training and evaluating whole‑body humanoid locomotion‑manipulation policies. Built on Isaac Sim and MuJoCo, it offers many robots, thousands of objects, dozens of indoor scenes, and 50+ tasks. It provides data‑generation pipelines, a client‑server evaluation setup with the Ψ₀ foundation model, and Docker/Nix/UV install options. The repo includes benchmark results and links to model and data on Hugging Face.

WebRPA is an open‑source, Windows‑only visual RPA suite that lets users build web, desktop and Android automation flows by dragging‑and‑dropping from a library of 571 modules. It ships with bundled Python 3.13 and Node.js runtimes, runs completely offline, and includes an AI assistant that can generate, debug and self‑heal workflows via OpenAI‑compatible APIs. Features include visual debugging, versioning, one‑click EXE packaging, optional heavy‑weight capability packs, and a modern React + FastAPI stack. Licensed under AGPL‑3.0 for non‑commercial use, with a commercial license available.

A Python package for aircraft and engineered system design and optimization using automatic differentiation and differentiable physics models.

A collection of simulation environments and demonstrations for Open-RMF, enabling interoperability and traffic management among diverse robot fleets in complex facilities.

A ROS wrapper for Intel RealSense cameras that enables depth and color sensor data to be published as ROS topics for robotic applications.

A high-fidelity robot simulation platform combining Unreal Engine 5 and MuJoCo to provide realistic rendering, high-rate physics, and ROS 2-compatible data streaming.

A robotic intelligence stack for the Unitree Go2 that enables autonomous navigation, dynamic obstacle avoidance, and semantic scene understanding using the Odin1 sensing module.

A Python SDK for the Piper robot arm that processes CAN bus data frames into custom data types for easier hardware control and monitoring.

T-Rex is a tactile-reactive dexterous manipulation framework that uses an asynchronous Mixture-of-Transformers to react dynamically to high-frequency touch signals for delicate robotic tasks.

An open-source project that converts basic robotic lawn mowers into GPS-guided, map-aware autonomous machines, removing the need for perimeter wires.

A software framework for real-time predictive control using MuJoCo, enabling the authoring and solving of complex robotics tasks through various planning methods.

A resilient, tightly-coupled multi-sensor SLAM framework that maintains accurate localization for wheeled, legged, and aerial robots even during sensor degradation or calibration errors.

An open-source L2 ADAS software stack that enables autonomous cruise control and lane keeping using a hybrid AI architecture and a single monocular camera without needing HD maps.

A heterogeneous architecture for robot reinforcement learning that decouples task definitions from physics simulators and hardware, enabling a seamless switch between backends like MuJoCo and IsaacGym.

A real-time physics simulation engine and SDK that provides tools for rigid body dynamics, destruction, and fluid simulation, including Python bindings for AI tensor interoperability.

An Unreal Engine 5-based simulation platform for developing and evaluating LLM/VLM agents in realistic physical and social environments.

A ROS2 package that provides a pre-integrated foundation for autonomous navigation, supporting physical robots (2WD, 4WD, Mecanum) and Gazebo simulations.

A C++ library for smoothing and mapping in robotics and vision that uses factor graphs and Bayes networks for state estimation and optimization.

A self-hosted web platform for running and managing multiple virtual iPhones in a browser using Apple's Virtualization.framework.

A GPU-accelerated elevation mapping framework for legged robots that fuses multi-modal data like geometry and semantics to create accurate real-time terrain maps.

Isaac Lab‑Arena is an alpha‑stage, open‑source extension to NVIDIA Isaac Lab that lets you build robot simulation environments by mixing reusable *scene*, *embodiment*, and *task* primitives. It assembles these at runtime, supports sequential task chaining, natural‑language object placement, and large‑scale parallel evaluation, making it easier to benchmark or train generalist robot policies. Installation works via the `uv` Python manager or a provided Docker image; the library is Linux‑only and requires an NVIDIA GPU and Isaac Sim. The project is under Apache 2.0 (with proprietary Isaac Sim components) and is intended for research rather than production.

A real-world reinforcement learning framework for robotic manipulation that enables a unified pipeline from imitation learning to offline/online RL and one-step policy distillation.

A reinforcement learning extension for Isaac Lab that enables humanoid robots, such as the Unitree G1, to learn complex motions via DeepMimic and Adversarial Motion Priors.

MobiAgent is an open‑source framework for running LLM‑based mobile agents on Android/HarmonyOS devices. It provides the MobiMind model family, an Agent Record‑&‑Replay system (AgentRR), and the MobiFlow benchmark. The repo includes on‑device inference support, multiple memory modules (user‑profile, experience, action), a multi‑task runner that controls phones via ADB, and a ready‑to‑install Android app. Detailed setup steps, model deployment instructions, and citation info are provided.

TinyMPC is a Model-Predictive Control solver designed to run efficiently on resource-constrained microcontrollers for real-time robotics control.

A C++ and Python library for autonomous agricultural vehicles that computes efficient, kinematically feasible coverage paths for non-convex fields with obstacles.

OpenExo is a control system for wearable robots that uses dual microcontrollers to manage exoskeleton hardware.

An AI-assisted visual programming IDE for hardware development that enables non-professional users to prototype projects across 100+ different development boards.

LeKiwi is a low-cost mobile manipulator featuring a holonomic drive and a robotic arm, designed for accessible robotics research and data collection.

A ROS 2 wrapper for OrbbecSDK that integrates Orbbec depth cameras and LiDAR devices into the ROS 2 ecosystem for robotics applications.

A Python-based model predictive controller for quadruped robots that offers both gradient-based and GPU-accelerated sampling-based optimization for stable locomotion.

OpenSim is a software framework for developing musculoskeletal models and creating dynamic simulations of movement to study human and animal biomechanics.

An open-source hardware platform providing CAD files, BOMs, and assembly guides for building and customizing robotic arms.

τ₀‑VLA is a hierarchical robot foundation model that pairs a Qwen 3.5 vision‑language backbone with a flow‑matching action expert. A memory‑augmented high‑level policy proposes sub‑tasks and, when needed, performs world‑model‑guided test‑time search; a low‑level policy executes the chosen sub‑task across robot embodiments via a unified 40‑dimensional state/action space. The repository includes training recipes, adapters for different robots, a serving stack, and checkpoints (including a LIBERO‑specific version).

An energy-based physics simulation framework for soft-bodied virtual creatures that uses automatic differentiation to compute forces and derivatives.

A set of rules and scripts that enable AI Agents to configure, build, flash, and debug TI MSPM0 microcontrollers by directly interacting with SysConfig and hardware toolchains.