A Lightweight Weighted Ensemble Approach for Sensor-Based Terrain Classification
Received: 24 December 2025 | Revised: 21 January 2026 and 25 January 2026 | Accepted: 28 January 2026 | Online: 6 February 2026
Corresponding author: Eren Yildirim
Abstract
The properties of a terrain on which a robot travels have a crucial impact on the design and function of the robot. Therefore, sensing the terrain type using sensors is essential. This study proposes a lightweight method based on time-series data collected by multiple inertial sensors located on the robot. The data in Cartesian coordinates are converted into cylindrical and spherical coordinates and then fused. Next, the average value of each feature over the measurement period is calculated for dimensionality reduction. Prior to classification, data augmentation is applied to the open-source dataset to increase the number of samples. The terrain types used in this study are concrete, grass, pebbles, sand, paving stone, and synthetic running track. Testing the proposed model with a Weighted Ensemble (WE) classifier yielded 97.37% accuracy. The findings showed that using multiple coordinate systems in the reduced feature set increases classification accuracy while maintaining a relatively low computational load.
Keywords:
terrain classification, data fusion, Weighted Ensemble (WE) classifierDownloads
References
Z. Zhou, G. Jiao, W. Chen, Y. Lu, and S. Luo, "Wave equivalent plane estimation for wave reconstruction in imaging through water waves," in 2017 IEEE International Conference on Real-time Computing and Robotics, Okinawa, Japan, 2017, pp. 459–464. DOI: https://doi.org/10.1109/RCAR.2017.8311905
L. Ojeda, J. Borenstein, G. Witus, and R. Karlsen, "Terrain characterization and classification with a mobile robot," Journal of Field Robotics, vol. 23, no. 2, pp. 103–122, Feb. 2006. DOI: https://doi.org/10.1002/rob.20113
M. G. H. Nampoothiri, B. Vinayakumar, Y. Sunny, and R. Antony, "Recent developments in terrain identification, classification, parameter estimation for the navigation of autonomous robots," SN Applied Sciences, vol. 3, no. 4, Mar. 2021, Art. no. 480. DOI: https://doi.org/10.1007/s42452-021-04453-3
P. Sarcevic et al., "Online Outdoor Terrain Classification Algorithm for Wheeled Mobile Robots Equipped with Inertial and Magnetic Sensors," Electronics, vol. 12, no. 15, July 2023, Art. no. 3238. DOI: https://doi.org/10.3390/electronics12153238
C. Weiss, N. Fechner, M. Stark, and A. Zell, "Comparison of Different Approaches to Vibration-based Terrain Classification," in 3rd European Conference on Mobile Robots, Freiburg, Germany, 2007. DOI: https://doi.org/10.1007/978-3-540-74764-2_1
E. Nebot, J. Guivant, and S. Worrall, "Haul truck alignment monitoring and operator warning system," Journal of Field Robotics, vol. 23, no. 2, pp. 141–161, 2006. DOI: https://doi.org/10.1002/rob.20114
T. Arafin, A. Hosen, Z. Najdovski, L. Wei, M. Rokonuzzaman, and M. Johnstone, "Advances and Trends in Terrain Classification Methods for Off-Road Perception," Journal of Field Robotics, vol. 42, no. 7, pp. 3515–3544, Oct. 2025. DOI: https://doi.org/10.1002/rob.22586
M. K. Villareal and A. F. Tongco, "Remote Sensing Techniques for Classification and Mapping of Sugarcane Growth," Engineering, Technology & Applied Science Research, vol. 10, no. 4, pp. 6041–6046, Aug. 2020. DOI: https://doi.org/10.48084/etasr.3694
H. Wang, E. Lu, X. Zhao, and J. Xue, "Vibration and Image Texture Data Fusion-Based Terrain Classification Using WKNN for Tracked Robots," World Electric Vehicle Journal, vol. 14, no. 8, Aug. 2023, Art. no. 214. DOI: https://doi.org/10.3390/wevj14080214
D. Csík, Á. Odry, J. Sárosi, and P. Sarcevic, "Inertial sensor-based outdoor terrain classification for wheeled mobile robots," in 2021 IEEE 19th International Symposium on Intelligent Systems and Informatics, Subotica, Serbia, 2021, pp. 159–164. DOI: https://doi.org/10.1109/SISY52375.2021.9582504
T. Knuth and P. Groves, "IMU Based Context Detection of Changes in the Terrain Topography," in 2023 IEEE/ION Position, Location and Navigation Symposium, Monterey, CA, USA, 2023, pp. 680–690. DOI: https://doi.org/10.1109/PLANS53410.2023.10140086
N. Erickson et al., "AutoGluon-Tabular: Robust and Accurate AutoML for Structured Data." arXiv, Mar. 13, 2020.
M. V. C. Aragão et al., "A practical evaluation of AutoML tools for binary, multiclass, and multilabel classification," Scientific Reports, vol. 15, no. 1, May 2025, Art. no. 17682. DOI: https://doi.org/10.1038/s41598-025-02149-x
M. Sátor-Érsek, S. Kajan, and L. Körösi, "Vibration based terrain classification for mobile robot using machine learning methods," Journal of Electrical Engineering, vol. 76, no. 5, pp. 420–428, Oct. 2025. DOI: https://doi.org/10.2478/jee-2025-0044
Downloads
How to Cite
License
Copyright (c) 2026 Eren Yildirim, Yucel Batu Salman
This work is licensed under a Creative Commons Attribution 4.0 International License.
Authors who publish with this journal agree to the following terms:
- Authors retain the copyright and grant the journal the right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) after its publication in ETASR with an acknowledgement of its initial publication in this journal.
