This repository contains a simple graph solver/optimizer for the 2D graph SLAM problem. This was done as part of MEEG 667-010 estimation class at the University of Delaware. The data was originally taken from Luca Carlone's preprocessed data. Originally the raw datasets where 2D LiDAR scans, but have been already converted into the g2o format for simplification. The solver is based on the A Tutorial on Graph-Based SLAM by Grisetti et al. Simply run the build_and_solve.m script to load the graph and optimize it. Figures below have the initial guess in blue with the optimized value in red.
