logo

Nonlinear Sensor Fusion Based on Support Vector Regression for Distance Estimation in Mobile Robotics

10.24050/reia.v22i44.1899
Published 2025-07-17
Articles

Abstract

Accurate distance estimation is a critical challenge in mobile robotics, particularly in dynamic environments where sensor uncertainties can impair navigation and obstacle avoidance. This paper proposes a nonlinear sensor fusion method based on Support Vector Regression (SVR) to enhance distance estimation using a low-cost setup comprising a Logitech C270 camera with ArUco fiducial markers and an HC-SR04 ultrasonic sensor, integrated via an ESP32 microcontroller and ROS2 middleware. The experimental setup involved collecting data from 20 to 400 cm with 10 cm intervals, incorporating controlled occlusions to simulate real-world conditions. The SVR model, optimized with a radial basis function kernel (C=10, gamma=0.5, epsilon=0.5), fuses visual and acoustic data to improve robustness and precision. Results from validation experiments (30-200 cm) show that the fused distance estimates achieve an average absolute error of 1.25 cm in normal conditions, with a peak mean squared error of 35.44 cm² at 30 cm, indicating limitations at short ranges. Under occlusion, the fusion mitigates errors from individual sensors (e.g., 19.81 cm for sonar, 35.08 cm for camera at 100 cm), reducing the average error to 32.49 cm. The main contributions include a cost-effective fusion approach that enhances distance estimation robustness in adverse conditions and outperforms single-sensor methods. Future work will explore deep learning techniques and additional sensors to address short-range and occlusion challenges, aligning with trends in multimodal sensor integration for autonomous navigation.

Keywords

Categories

Authors

  • Santiago Alexander Durán-Puentes

    Colombia

  • Gustavo Alonso Acosta-Amaya

    Colombia

  • Deimer Andres Miranda-Montoya

    Sede Medellín

  • Jovani Alberto Jimenez-Builes

    Sede Medellín

    https://orcid.org/0000-0001-7598-7696

References

Alatise, M.; Hancke, G. (2020). A review on challenges of autonomous mobile robot and sensor fusion methods. IEEE Access, 8, 39830-39846. https://doi.org/10.1109/ACCESS.2020.2975643

Andronie, M.; Lăzăroiu, G.; Iatagan, M.; Hurloiu, I.; Ștefănescu, R.; Dijmărescu, A.; Dijmărescu, I. (2023). Big data mana-gement algorithms, deep learning-based object detection technologies, and geospatial simulation and sensor fusion tools in the internet of robotic things. ISPRS International Journal of Geo-Information, 12(35). https://doi.org/10.3390/ijgi12020035

Barreto, A.; Gómez, A.; Escobedo, J.; Cuan, E.; Cruz, S. (2022). Sensor data fusion for a mobile robot using neural net-works. Sensors, 22(1), 305. https://doi.org/10.3390/s22010305

Duan, S.; Shi, Q.; Wu, J. (2022). Multimodal sensors and ML-based data fusion for advanced robots. Advanced Intelligent Systems, 4(12), 2200213. https://doi.org/10.1002/aisy.202200213

Fayyad, J.; Jaradat, M.; Gruyer, D.; Najjaran, H. (2020). Deep learning sensor fusion for autonomous vehicle perception and localization: A review. Sensors, 20(15), 4220. https://doi.org/10.3390/s20154220

Hafez, O.; Joerger, M.; Spenko, M. (2024). How safe is particle filtering-based localization for mobile robots? An inte-grity monitoring approach. IEEE Transactions on Robotics, 40, 3372–3387. https://doi.org/10.1109/TRO.2024.3420798

Housein, A. A.; Xingyu, G.; Li, W.; Huang, Y. (2022). Extended Kalman filter sensor fusion in practice for mobile robot localization. International Journal of Advanced Computer Science and Applications, 13(2), 33-38. https://doi.org/10.14569/IJACSA.2022.0130204

Kaczmarek, A.; Rohm, W.; Klingbeil, L.; Tchórzewski, J. (2022). Experimental 2D extended Kalman filter sensor fusion for low-cost GNSS/IMU/Odometers precise positioning system. Measurement, 193, 110963. https://doi.org/10.1016/j.measurement.2022.110963

Krishnamurthi, R.; Kumar, A.; Gopinathan, D.; Nayyar, A.; Qureshi, B. (2020). An overview of IoT sensor data proces-sing, fusion, and analysis techniques. Sensors, 20(21), 6076. https://doi.org/10.3390/s20216076

Kumari, N.; Kulkarni, R.; Ahmed, M.; Kumar, N. (2021). Use of Kalman filter and its variants in state estimation: A re-view. En: Artificial Intelligence for a Sustainable Industry 4.0 (pp. 213-230). https://doi.org/10.1007/978-3-030-77070-9_13

Li, C.; Fahmy, A.; Sienz, J. (2019). An augmented reality based human-robot interaction interface using Kalman filter sensor fusion. Sensors, 19(20), 4586. https://doi.org/10.3390/s19204586

Li, Y.; He, L.; Zhang, X.; Zhu, L.; Zhang, H.; Guan, Y. (2019). Multi-sensor fusion localization of indoor mobile robot. En: IEEE International Conference on Real-time Computing and Robotics (RCAR) (pp. 481-486). https://doi.org/10.1109/RCAR47638.2019.9044006

Lv, M.; Wei, H.; Fu, X.; Wang, W.; Zhou, D. (2022). A loosely coupled extended Kalman filter algorithm for agricultural scene-based multi-sensor fusion. Frontiers in Plant Science, 13, 849260. https://doi.org/10.3389/fpls.2022.849260

Moreira, A.; Costa, P.; Lima, J. (2020). New approach for beacons based mobile robot localization using Kalman filters. Procedia Manufacturing, 51, 512-519. https://doi.org/10.1016/j.promfg.2020.10.072

Mráz, E.; Trizuljak, A.; Rajchl, M.; Sedláček, M.; Štec, F.; Stanko, J.; Rodina, J. (2024). Multi-sensor fusion for robust in-door localization of industrial UAVs using particle filter. Journal of Electrical Engineering, 75(4), 304-316. https://doi.org/10.2478/jee-2024-0037

Oliveira, R.; Ott, L.; Guizilini, V.; Ramos, F. (2020). Bayesian optimisation for safe navigation under localisation uncer-tainty. En: Robotics Research: The 18th International Symposium ISRR (pp. 489-504). Springer International Publis-hing. https://doi.org/10.48550/arXiv.1709.02169

Ortega, J.; Gutiérrez, O.; Aguirre, J. (2022). Simulación del control de un robot móvil tipo Rover basado en fusión de sensores mediante filtro de partículas. Pädi Boletín Científico de Ciencias Básicas e Ingenierías del ICBI, 10, 100-106. https://doi.org/10.29057/icbi.v10iEspecial6.9366

Sander, J.; Beyerer, J. (2013). Bayesian fusion: Modeling and application. En: 2013 Workshop on Sensor Data Fusion: Trends, Solutions, Applications (SDF) (pp. 1-6). IEEE. https://doi.org/10.1109/SDF.2013.6698254

Shaukat, N.; Ali, A.; Javed, M.; Moinuddin, M.; Otero, P. (2021). Multi-sensor fusion for underwater vehicle localization by augmentation of RBF neural network and error-state Kalman filter. Sensors, 21(4), 1149. https://doi.org/10.3390/s21041149

Shen, Y.; Liu, H.; Liu, X.; Zhou, W.; Zhou, C.; Chen, Y. (2024). Localization through particle filter powered neural network estimated monocular camera poses. arXiv preprint, arXiv:2404.17685. https://doi.org/10.48550/arXiv.2404.17685

Tang, Q.; Liang, J.; Zhu, F. (2023). A comparative review on multi-modal sensors fusion based on deep learning. Signal Processing, 213, 109165. https://doi.org/10.1016/j.sigpro.2023.109165

Urrea, C.; Agramonte, R. (2021). Kalman filter: Historical overview and review of its use in robotics 60 years after its creation. Journal of Sensors, 2021(1), 9674015. https://doi.org/10.1109/MSP.2025.3569395

Yeong, D.; Velasco, G.; Barry, J.; Walsh, J. (2021). Sensor and sensor fusion technology in autonomous vehicles: A re-view. Sensors, 21(6), 2140. https://doi.org/10.3390/s21062140

Statistics

Abstract
0
PDF accesses
0

Dimensions

PlumX

Downloads

Download data is not yet available.