Implementation of Fast Spectral Scan Matching (F-SSM) and SVM-based place recognition for coarse-to-fine global localization of a mobile robot, based on Park & Roh (2016). Built and tested in the NEU Racing Gymnasium environment using 2D lidar scan data.
python numpy svm particle-filter lidar ransac place-recognition mobile-robotics gymnasium robot-localization spectral-methods mobile-robot scan-matching range-scan neu-racing
-
Updated
Apr 18, 2026 - Jupyter Notebook