An AI‑driven Bitcoin futures trading bot for OKX that uses DeepSeek‑chat to generate buy/hold/sell signals, places real stop‑loss/take‑profit orders, and provides a Streamlit web dashboard for live monitoring. Deployable via Baota panel, Docker, or plain Python.

A ROS2 adaptation of FASTLIO2 that adds loop closure, relocalization, and consistency map optimization for LiDAR-Inertial Odometry.

A low-cost, open-source brain-computer interface platform featuring 8-channel BLE hardware and SDKs for Python and Android.

A PyTorch‑based library that provides exact second‑order (Levenberg‑Marquardt, Schur‑complement) optimizers for sparse robotics problems such as Bundle Adjustment and Pose‑Graph Optimization, with CUDA acceleration, PyPose integration, and support for torch.compile.

An imitation learning benchmark for whole-body control featuring diverse humanoid and quadruped environments with extensive motion capture datasets.

A low-cost, open-source LiDAR-Inertial-Visual fusion perception kit for robotics and embodied AI, providing hardware designs and software drivers for 3D spatial awareness.

A real-time Nonlinear Model Predictive Controller (NMPC) for humanoid robots that optimizes full-order torque-level dynamics for loco-manipulation.

A ROS 2 integration of libfranka that allows for efficient control and development of Franka Robotics research arms and mobile platforms.

PyRoboSim is a 2D mobile robot simulator for behavior prototyping that optionally integrates with ROS 2.

A robust C implementation of the Cyphal/CAN transport layer for high-integrity real-time embedded systems in aerospace and robotics.

A real-time Visual-Inertial SLAM framework that replaces traditional handcrafted features with SuperPoint and LightGlue deep learning models to handle challenging imaging conditions.

A PyTorch-based package providing reinforcement learning implementations for continuous control tasks in the MuJoCo physics simulator.

A set of C# libraries and tools that enable .NET applications and Unity projects to communicate with ROS nodes for robotics simulations and automation.

A type-safe Rust library for rigid body transforms and spatial math, simplifying coordinate conversions between systems like WGS84, ECEF, and NED.

A set of libraries providing a high-level, cross-platform API for MAVLink, enabling developers to control drones and unmanned vehicles across multiple programming languages.

A C++ framework for optimizing graph-based nonlinear error functions, commonly used for SLAM and Bundle Adjustment in robotics and computer vision.

A software development kit for Boston Dynamics' Spot robot that enables developers to build custom applications and payloads to control the robot and read sensor data.

A scalable Reinforcement Learning framework specialized for Physical AI applications, providing a decoupled architecture for large-scale training workloads.

RFUAV provides a publicly‑available RF‑signal dataset for UAV detection/identification, along with Python/MATLAB tools to turn raw IQ recordings into spectrograms, train vision‑style classifiers (ResNet, ViT, etc.) or YOLOv5 detectors, and evaluate performance across a wide SNR range.

SMACC2 is a C++ library for building real‑time, event‑driven state machines on ROS 2, targeting complex multi‑component robots. It offers orthogonal regions, compile‑time validation, ready‑made ROS 2 client wrappers, and a library of example state machines, with full Doxygen documentation and a free runtime visualiser.

SpatialBot is a Vision-Language-Action model that enhances spatial understanding and depth perception in VLMs, enabling precise spatial reasoning and robot manipulation.

A mobile GUI agent framework for Android that enables AI agents to perceive and operate app interfaces on real devices for automated, long-running workflows.

A Python developer kit for accessing and working with the NVIDIA Physical AI Autonomous Vehicles Dataset to build end-to-end driving systems.

A Lie theory library for robotics state-estimation that provides C++ and Python tools for handling rotations, translations, and velocities across various Lie groups.

A pure Python library for estimating the orientation of mobile systems using various attitude and heading reference system (AHRS) algorithms.

rviz is a 3D visualizer for the Robot Operating System (ROS) framework, allowing developers to see robot data in 3D.

WorldEngine is an open‑source post‑training platform for autonomous driving that automatically discovers rare, safety‑critical scenarios from real logs, reconstructs them with photorealistic 3‑D Gaussian Splatting, and fine‑tunes driving models via RL on the generated roll‑outs. It includes two subsystems—SimEngine for closed‑loop simulation and AlgEngine for training/evaluation—showing large gains on rare‑case benchmarks and a real‑world deployment with zero disengagements.

Evo-1 is a lightweight Vision-Language-Action (VLA) model for robotic manipulation that translates visual and textual instructions into robot actions across various simulations and real hardware.

UniWM is a unified, memory-augmented world model for visual navigation that integrates egocentric visual foresight and planning within a single multimodal autoregressive backbone.

A flexible AI agent framework for developing and deploying Embodied AI features, enabling robots to use generative AI for reasoning, multi-modal interaction, and complex task execution.

An HTTP server that streams ROS image topics to web browsers in multiple formats, enabling easy remote monitoring of robot camera feeds.

A lightweight stereo visual-inertial odometry system that uses statistical uncertainty learning and sliding window optimization for real-time robot state estimation.

A collection of ROS-based IMU filters and visualizers that fuse angular velocity, acceleration, and magnetic readings into a stable orientation estimate.

DeepThinkVLA is a Vision‑Language‑Action model that inserts an explicit chain‑of‑thought reasoning step before generating robot actions. Using a hybrid decoder (autoregressive reasoning → bidirectional parallel action generation), the authors train on a newly curated embodied CoT dataset and then refine with outcome‑driven reinforcement learning. The model achieves 97 % average success on the LIBERO benchmark, runs at only 0.175× the latency of a pure autoregressive baseline, and transfers zero‑shot to the newer LIBERO Plus suite.

A real-time RGB-D inertial odometry system designed for resource-constrained robots to maintain accurate localization in dynamic environments by filtering out moving objects.

A collection of Unity ML-Agents experiments for training reinforcement learning agents in vehicle control, hovering, and aerial ball handling.

Colette is a self‑hosted RAG platform that indexes PDFs as images, preserving tables, figures and layout. It combines vision embeddings, BM25 lexical search, and optional HyDE prompting, and runs on high‑end GPUs via Docker. Ideal for confidential technical documentation.

A PyTorch and Hugging Face native toolkit for the reproducible training and evaluation of imitation learning policies for robotic desktop manipulation.

A reinforcement learning framework for legged robots that integrates multiple simulators (IsaacGym, Genesis, IsaacSim) and implements various state-of-the-art locomotion methods.

A C++ library for communicating with Invensense MPU-6500 and MPU-9250/9255 IMUs, providing easy access to motion and magnetic data via I2C or SPI.

ROCm / legacy‑rocm‑build is the (now‑deprecated) repository that assembles AMD’s open‑source GPU compute stack—drivers, HIP runtime, math libraries, profiling tools, and more—so developers can run AI/ML workloads on AMD GPUs. It points users to install guides and to the newer TheRock monorepo.

An open-source Warehouse Execution System (WES) that provides real-time orchestration and AI-powered optimization for automated warehouses using robots and IoT.

An open-source co-simulation framework for Cooperative Driving Automation (CDA) that integrates CARLA and SUMO to research multi-vehicle cooperation and V2X communication.

GenSim is an LLM-powered pipeline that automatically generates robotic simulation tasks and environments to augment training data for multitask robotic policies.

A comprehensive survey and taxonomy of Embodied Co-Design (ECD), focusing on the joint optimization of an agent's morphology and control to create more capable intelligent robots and virtual creatures.

A JEPA-inspired Physical World Model for robots that uses an Energy-Based Transformer to simulate and refine action trajectories based on physical consistency.

RStan is the R interface to Stan, a probabilistic programming language for Bayesian inference. It lets R users write Stan models, compile them, and run state‑of‑the‑art MCMC or variational inference, returning results as R objects that integrate with standard diagnostic and plotting tools.

A ROS 2 package that provides tools to create, configure, build, and flash micro-ROS firmware for various embedded platforms and microcontrollers.