Tracking animal movement is an important aspect of neuroscience research, particularly in studies wishing to investigate the relationships between behavioral changes and neural activity. To achieve this, treadmill systems are commonly utilized to provide controlled environments for monitoring animal locomotion. However, existing treadmill systems rely on measuring only the rotational speed of its motors or belt, failing to capture changes in an animal's actual position and velocity. This limitation is problematic for neural correlation studies, where accurate measurements of animal movement are essential. To address these concerns, we present a novel modular tracking system that enables real-time monitoring of animal position and velocity during treadmill-based rodent experiments at a sampling rate of 10 Hz. This system comprises of interlocking, 3D-printed modules equipped with infrared (IR) break-beam sensors, allowing customizable configurations for diverse experimental setups. Additionally, it supports both digital and analog output triggers, for integration with neural recording systems to synchronize data acquisition of treadmill, behavioral, and neural data. In this study, we validate our system by comparing position and velocity data collected from mice moving on a treadmill at 0.1 m/s against measurements obtained from DeepLabCut, a leading machine learning-based pose-estimation tool. The results of our investigation demonstrate strong correlations between our system's real-time position and velocity measurements against post-processed video tracking data with an average Spearman's rank correlation of r = 0.8 for each. Overall, this versatile system provides an accurate and accessible retrofit solution for tracking animal movement in controlled neuroscience experiments, enhancing the precision of behavioral analysis.Clinical Relevance-The described IR system offers a practical solution for researchers who study movement disorders in rodent models, as it provides precise, real-time locomotor tracking without the need for complex computer vision analyses.