Open-source alternatives to FAST LIO

Open_vins is a visual-inertial odometry framework and SLAM system designed for robotic state estimation. It uses an Extended Kalman Filter to fuse high-frequency inertial sensor data with visual feature tracks to estimate the position and orientation of a moving device. The system features a sensor calibration suite for calculating intrinsic and extrinsic parameters, as well as temporal offsets between cameras and inertial measurement units. It includes a manifold interpolator that uses B-Spline curves over the special Euclidean group to produce smooth trajectory paths between discrete pose e

VINS-Mono is a monocular visual-inertial odometry system and loop closure SLAM framework. It functions as a real-time state estimator that fuses data from a single camera and an inertial measurement unit to determine a robot's position and orientation. The project includes a non-linear optimizer for robotics and tools for sensor calibration. The system distinguishes itself through online sensor calibration, which automatically determines spatial extrinsics and temporal offsets between the camera and inertial unit during operation. It also incorporates rolling shutter distortion correction to

filterpy is a toolkit for Bayesian state estimation, Gaussian statistical analysis, and time-series noise reduction. It provides a library of linear and non-linear Kalman filters, as well as routines for non-Gaussian state estimation and signal smoothing. The project implements a variety of estimation methods, including particle filtering using Markov Chain Monte Carlo and resampling, and discrete Bayes filtering. It also includes a suite of algorithms for refining historical state estimates through backward and fixed-lag smoothing. Additional capabilities cover multivariate Gaussian analysi

This project is an educational resource and toolkit for implementing Bayesian estimation and Kalman filters in Python. It provides a framework for constructing linear and non-linear filters to estimate the state of dynamic systems by combining noisy sensor data with mathematical process models. The library focuses on probabilistic state estimation, utilizing recursive Bayesian updating and state-space mathematical modeling to refine beliefs about system states. It includes utilities for simulating dynamic systems, allowing users to generate synthetic trajectories and sensor observations to va

The Point Cloud Library is a collection of C++ algorithms designed for filtering, registering, and analyzing large-scale 3D spatial datasets. It provides a framework for 3D point cloud processing, incorporating tools for spatial data filtering and geometric feature estimation. The library includes specialized systems for aligning multiple spatial datasets into a single unified coordinate system and a rendering engine for the visual inspection and analysis of processed point cloud data. It also features tools for calculating spatial descriptors to identify structural patterns and shapes within

PythonRobotics is a comprehensive collection of modular robotics algorithms and educational simulations designed for autonomous navigation, state estimation, and motion control. The project provides a library of standalone implementations for path planning, localization, mapping, and kinematics, serving as a resource for researchers and students to experiment with foundational and advanced robotic theories. The project distinguishes itself through an algorithm-centric design where each module functions as an isolated script, allowing for independent testing and clear pedagogical demonstration

SensorsCalibration is a toolkit for computing the intrinsic and extrinsic parameters of cameras and LiDAR sensors in automotive environments. It provides tools for calculating internal camera focal properties, aligning LiDAR point clouds to 2D camera images, and determining the spatial transformations between multiple sensors and the vehicle coordinate system. The project includes a targetless calibration framework that aligns sensors to a vehicle by analyzing natural features in road environments without the need for specialized physical markers. It also supports factory calibration processe

Cartographer is a cross-platform robotics library and framework for simultaneous localization and mapping in 2D and 3D spaces. It functions as a real-time mapping engine that constructs environmental maps while tracking a device's position and orientation using continuous sensor data processing. The system implements real-time SLAM to generate precise maps for autonomous navigation. It utilizes a localization system that determines a device's state within a mapped environment across different hardware platforms and sensor configurations. The framework covers spatial estimation through non-li

CGAL is a software library that provides a comprehensive collection of computational geometry algorithms and data structures. It is built around a geometry kernel that defines fundamental geometric primitives and operations, enabling the construction of complex geometric objects and the computation of geometric predicates with exact arithmetic for reliable results. The library covers a wide range of geometric computation capabilities, including the construction of convex hulls, triangulations of point sets, and the generation of Voronoi diagrams. It also supports the processing of polygonal m

MathUtilities is a collection of specialized toolkits providing engines for geometry, computer vision, mathematics, physics simulation, and signal processing. It functions as a comprehensive mathematics and physics library focused on linear algebra, numerical optimization, and geometric calculations for technical applications. The project distinguishes itself through a physics simulation toolkit and a 3D geometry engine. These provide capabilities for Verlet integration, iterative inverse kinematics solvers, distance field rendering via volumetric raymarching, and mesh geometry deformation. I

MotionPlanning is a software library designed for autonomous navigation, providing a comprehensive suite of tools for path planning, trajectory generation, and vehicle control. It enables the calculation of collision-free routes and dynamic movement paths for autonomous vehicles operating in complex, changing environments. The project distinguishes itself by integrating hierarchical motion decomposition, which separates high-level route planning from low-level trajectory generation to manage computational complexity. It employs kinematic bicycle modeling and trailer dynamics simulation to ens

This project is a multi-object tracking library and computer vision toolkit designed to maintain consistent identity IDs for objects across video frames. It provides a motion-based object tracking system that converts raw detections into stable temporal tracks, enabling the analysis of object movement and behavior over time. The toolkit distinguishes itself through advanced identity maintenance, utilizing Kalman filters for linear motion tracking and sparse optical flow for camera motion estimation. It features multi-stage object association to recover occluded objects and non-linear motion t

VINS-Fusion is a multi-sensor fusion framework and visual-inertial odometry system. It integrates camera images, inertial measurement unit data, and global positioning signals through a non-linear optimization system to track the position and orientation of autonomous vehicles. The system includes a visual loop closure engine that utilizes a bag-of-words approach to recognize previously visited locations and correct trajectory drift. It further provides tools for online spatio-temporal calibration to determine the physical offset and time synchronization between cameras and inertial sensors d

OM1 is a multimodal AI agent runtime and orchestration framework designed to connect large language models to physical robot hardware and sensors. It provides an execution environment that processes audio, video, and sensor data to drive autonomous decisions and actions in real-world settings. The system integrates a robotics SLAM and navigation stack with a hardware abstraction layer, allowing high-level AI commands to be translated into low-level motor and actuator instructions. It distinguishes itself by incorporating blockchain-based governance to enforce immutable operational rules and p

PX4-Autopilot is a professional-grade flight control software stack designed for autonomous unmanned vehicles, including multicopters, fixed-wing aircraft, and vertical takeoff and landing platforms. It operates as a modular, real-time framework that decouples flight control logic from hardware drivers through a publish-subscribe middleware architecture. The system utilizes a deterministic microkernel runtime to execute time-critical flight control loops and sensor fusion tasks, ensuring stable navigation and vehicle operation. The platform distinguishes itself through a parameter-driven conf

jetson-inference is a set of libraries and tools for executing optimized deep learning models on embedded GPU hardware. Its primary purpose is to enable real-time computer vision and AI inference at the edge with low latency and high throughput. The project distinguishes itself through high-performance streaming analytics and the ability to execute concurrent AI pipelines on auto-grade silicon. It provides specialized support for multi-sensor stream processing, utilizing zero-copy data transport to load camera frames directly into GPU memory. The codebase covers a broad surface of capabiliti

Relativty is an open-source virtual reality headset kit that can be built for around $200 using DIY hardware and firmware, designed to work with SteamVR. At its core, it provides firmware for a 3-DoF VR headset that handles sensor calibration, orientation tracking, and communication with a PC over USB, along with a custom SteamVR-compatible driver that connects the self-built headset to the SteamVR ecosystem for launching games. The project distinguishes itself with an experimental full-body tracking module that uses a webcam and a neural network on an NVIDIA GPU to estimate 3D body position,

This research framework provides a deep learning driving simulator and a multimodal data pipeline for autonomous vehicle research. It centers on the creation of synchronized autonomous vehicle datasets, which combine high-frequency vehicle telemetry with camera frames to train neural networks. The project implements a convolutional neural network trainer specifically designed to predict steering angles and vehicle transition states from visual data. It features generative capabilities, using autoencoders and transition models to synthesize driving environments and simulate future vehicle move

mmcv is a foundation library for computer vision based on PyTorch. It provides a comprehensive system for constructing convolutional neural networks, a toolkit for image and video preprocessing, and a collection of high-performance deep learning vision operators. The project is distinguished by its hardware-accelerated kernels for complex operations such as deformable convolutions and region pooling. It features a configuration-driven framework that allows for the dynamic instantiation of network layers and the registration of custom modules without modifying code. The library covers a broad

QuantResearch is a quantitative research framework and specialized toolkit for algorithmic simulation, financial time-series analysis, and systematic trading. It provides an event-driven backtesting environment for validating strategies against historical tick and bar data, alongside a dedicated portfolio optimization engine for calculating asset weights and risk metrics. The project distinguishes itself through a machine learning finance toolkit that implements recurrent neural networks for price prediction and reinforcement learning for derivative pricing. It also features advanced statisti

pyslam is a framework for Simultaneous Localization and Mapping that combines Python flexibility with C++ performance. It is a sparse SLAM implementation designed to map environment geometry and track device location by processing image frames into 3D points. The project features a bridge for exposing high-performance C++ classes to Python scripts using zero-copy memory sharing. This integration allows for switching between a scripting interface for rapid prototyping and a compiled core for execution speed. The system includes a spatial map optimizer to refine 3D point and camera pose estima

Point-e is a system for 3D model synthesis that generates three-dimensional point clouds from natural language descriptions and two-dimensional images. It utilizes diffusion models to synthesize these spatial representations based on text prompts or source images. The project includes specialized tools for refining these outputs, such as a point cloud upsampler to increase the density and resolution of low-resolution models. It also provides a mesh converter that uses distance function regression to transform raw point cloud data into structured 3D meshes. The broader capability surface cove

Remove then revert (IROS 2020)

Implementation for PolarNet: An Improved Grid Representation for Online LiDAR Point Clouds Semantic Segmentation (CVPR 2020)

Open3D is a 3D data processing library, visualization engine, and machine learning library. It provides a framework for manipulating point clouds and meshes through specialized algorithms designed for 3D data science workflows. The project includes a toolkit for 3D scene reconstruction to generate spatial models and align surfaces from raw data. It also functions as a GPU accelerated framework that offloads intensive spatial computations to the graphics processor to increase processing speed. The library covers a broad range of capabilities including physically based light simulations for vi

LiDAR point cloud ground filtering / segmentation (bare earth extraction) method based on cloth simulation