Localization approaches for autonomous racing, such as pose-graph based Simultaneous Localization and Mapping (SLAM) [1] and Monte-Carlo Localization (MCL)-based (also called Particle Filtering, or PF) methods [2-4] depend on both exteroceptive and proprioceptive inputs. For example, LiDAR sensors offer range measurements for exteroceptive sensing, enabling the robot to perceive its environment. In contrast, proprioceptive measurements provide insight into the robot's internal states, processing signals from IMUs and wheel-odometry. Consequently, SLAM algorithms can map environments while localizing the robot. On the other hand, MCL-based techniques, relying on both sensing modalities and a pre-existing map, Figure 1: Comparison of poses generated by diff-drive [2] and solely determine the robot localization using MCL [3, 4].