scan-matching

A collection of GICP-based fast point cloud registration algorithms

Efficient and parallel algorithms for point cloud registration [C++, Python]

Point cloud registration pipeline for robot localization and 3D perception

Multi-threaded and SSE friendly NDT algorithm

K-Closest Points and Maximum Clique Pruning for Efficient and Effective 3-D Laser Scan Matching (RA-L 2022)

Laser scan matcher ported to ROS2

ROS package for NDT-PSO, a 2D Laser scan matching algorithm for SLAM

Globally localise your 2D LIDAR in a 2D map in no time [IROS'24]

Obtain robust odometry from your noisy panoramic 2D LIDAR [IROS'22]

The Fourier Scan Matcher: a correspondenceless and closed-form matching algorithm for range scans of 2D panoramic LIDAR sensors

This repository contains solution for SLAM lectures taught by Claus Brenner on YouTube.

Simple 2D point-to-point scan matcher implemented in Python. Works with ROS1.

Build and refine AI trading agents that self-improve prompts using market feedback to optimize performance without GPU requirements.

3-DoF pose from a single 2D LiDAR scan with no initial guess — ROS 2 global localization by FFT correlation of oriented (wall-normal) fields in image space

LiDAR Odometry with Scan Matching

[ROS package] Lidar odometry from panoramic 2D range scans. Method: scan-matching without using correspondences, based on properties of the Discrete Fourier Transform

GPU-accelerated ROS 2 LiDAR localization using CUDA ICP for real-time map-based 6-DoF pose estimation.

Self-Driving Car Engineer Nanodegree program, Course 4 - Localization Project