Statistical Machine Learning-based Odometry Correction for Cost-effective Indoor Robot Localization
Keywords:
ArUco marker-based localization, Indoor robot localization, Low-cost navigation systems, Multiple linear regression for robotics, Sensor fusion without LiDAR, Statistical error modelling, Wheel odometry drift correctionAbstract
Existing techniques for autonomous navigation, such as simultaneous localization and mapping (SLAM) using LiDAR, are economically expensive and computationally limiting for lightweight robotic systems. This study presents a low-cost method for an indoor mobile robot’s autonomous navigation that can be used as a cheaper alternative. This approach presents a mixed localization framework that uses differential drive kinematics, statistical machine learning, and sparse computer vision. The analytically derived surface data is used to train a multivariate linear regression model (MLR) to interpret raw dual-wheel optical encoder telemetry into corrected Cartesian displacement vectors. By keeping track of statistical variance and utilizing a dynamically growing error threshold, the system automatically limits the uncertainty of localization. When the error limit is reached, the system signals the monocular camera to scan for predefined ArUco markers mappers with the JSON-based coordinate database. Decoding these markers provides an absolute location reset for the robot, which instantly collapses the statistical uncertainty to zero. Physical models and simulations demonstrate that this method of localization significantly reduces the cumulative trajectory drift compared to raw odometry, offering a strong, highly computationally efficient alternative for autonomous navigation in an indoor environment.
References
R. Siegwart, I. R. Nourbakhsh, and D. Scaramuzza, Introduction to Autonomous Mobile Robots, 2nd ed. Cambridge, MA, USA: MIT Press, 2011.
F. Rubio, F. Valero, and C. Llopis-Albert, “A review of mobile robots: Concepts, methods, theoretical framework, and applications,” Int. J. Adv. Robot. Syst., vol. 16, no. 2, pp. 1–22, 2019.
J. Borenstein and L. Feng, “Measurement and correction of systematic odometry errors in mobile robots,” IEEE Trans. Robot. Autom., vol. 12, no. 6, pp. 869–880, Dec. 1996.
A. Martinelli, “The odometry error of a mobile robot with a synchronous drive system,” IEEE Trans. Robot., vol. 18, no. 3, pp. 399–405, Jun. 2002.
S. Garmpis, “Robot path estimation from odometry measurements: Loop closure based on autocorrelation,” Robot. Autom. Eng. J., vol. 5, no. 2, pp. 1–6, 2022.
X. Teng, “Multi-sensor fusion-based wheeled robot research on indoor positioning method,” Results Eng., vol. 21, 2024.
Y. Zhang, P. Shi, and J. Li, “3D LiDAR SLAM: A survey,” Photogramm. Rec., vol. 39, no. 187, pp. 1–20, 2024.
S. Chen, “The Gaussian distribution: Derivation method and its applications in selected examples,” Theor. Nat. Sci., vol. 38, pp. 1–10, 2024.
D. J. Olive, L. C. R. P. Watagoda, and H. S. R. A. Don, “Visualizing and testing the multivariate linear regression model,” Int. J. Stat. Probab., vol. 4, no. 1, pp. 126–136, 2015.
T. Hodson, T. Over, and S. Foks, “Mean squared error, deconstructed,” J. Adv. Model. Earth Syst., vol. 13, 2021.
S. Garrido-Jurado, R. Muñoz-Salinas, F. J. Madrid-Cuevas, and M. J. Marín-Jiménez, “Automatic generation and detection of highly reliable fiducial markers under occlusion,” Pattern Recognit., vol. 47, no. 6, pp. 2280–2292, Jun. 2014.
F. J. Romero-Ramirez, R. Muñoz-Salinas, and R. Medina-Carnicer, “Speeded up detection of squared fiducial markers,” Image Vis. Comput., vol. 76, pp. 38–47, Aug. 2018.
