{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,1,12]],"date-time":"2026-01-12T11:30:39Z","timestamp":1768217439955,"version":"3.49.0"},"reference-count":23,"publisher":"MDPI AG","issue":"2","license":[{"start":{"date-parts":[[2026,1,10]],"date-time":"2026-01-10T00:00:00Z","timestamp":1768003200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"Ministry of Educatio"},{"name":"Gyeongsangnam-do Provincial Government, Republic of Korea","award":["2025-RISE-16-007-0002"],"award-info":[{"award-number":["2025-RISE-16-007-0002"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Sensors"],"abstract":"<jats:p>Accurately estimating the absolute position of robots under external loads is challenging due to nonlinear dynamics, posture-dependent manipulability, and structural sensitivities. This study investigates a data-driven approach for absolute position prediction of a 2-DOF planar delta robot by learning time-series force signals generated during manipulability-driven free motion. Constant torques of opposite directions were applied to the robot without any position or trajectory control, allowing the mechanism to move naturally according to its configuration-dependent manipulability. Reaction forces measured at the end-effector and relative encoder variations were collected across a grid of workspace locations and used to construct a 12-channel time-series input. A hybrid deep learning architecture combining 1D convolutional layers and a bidirectional LSTM network was trained to regress the robot\u2019s absolute X\u2013Y position. Experimental results demonstrate that the predicted trajectories closely match the measured paths in the workspace, yielding overall RMSE values of 3.81 mm(X) and 2.94 mm(Y). Statistical evaluation using RMSE shows that approximately 83.73% of all test sequences achieve an error below 5 mm. The findings confirm that LSTM models can effectively learn posture-dependent dynamic behavior and force-manipulability relationships.<\/jats:p>","DOI":"10.3390\/s26020470","type":"journal-article","created":{"date-parts":[[2026,1,12]],"date-time":"2026-01-12T08:20:37Z","timestamp":1768206037000},"page":"470","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["LSTM-Based Absolute Position Estimation of a 2-DOF Planar Delta Robot Using Time-Series Data"],"prefix":"10.3390","volume":"26","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-4160-0868","authenticated-orcid":false,"given":"Seunghwan","family":"Baek","sequence":"first","affiliation":[{"name":"Department of Electrical and Electronic Engineering, Youngsan University, Yangsan 50510, Republic of Korea"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1968","published-online":{"date-parts":[[2026,1,10]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"216","DOI":"10.1007\/s10462-025-11223-9","article-title":"A comprehensive survey of deep learning for time series forecasting: Architectural diversity and open challenges","volume":"58","author":"Kim","year":"2025","journal-title":"Artif. Intell. Rev."},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"92306","DOI":"10.1109\/ACCESS.2024.3422528","article-title":"Deep Learning Models for Time Series Forecasting: A Review","volume":"12","author":"Li","year":"2024","journal-title":"IEEE Access"},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"5079","DOI":"10.1007\/s13042-025-02560-w","article-title":"Deep learning for time series forecasting: A survey","volume":"16","author":"Kong","year":"2025","journal-title":"Int. J. Mach. Learn. Cybern."},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"650","DOI":"10.1016\/j.procir.2021.03.088","article-title":"A survey on long short-term memory networks for time series prediction","volume":"99","author":"Lindemann","year":"2021","journal-title":"Procedia CIRP"},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"71","DOI":"10.1007\/s12530-025-09703-y","article-title":"Investigating LSTM-based time series prediction using dynamic systems measures","volume":"16","author":"Mahmoudi","year":"2025","journal-title":"Evol. Syst."},{"key":"ref_6","first-page":"158222","article-title":"Incidence of gastroenteropancreatic neuroendocrine tumours: A systematic review of the literature","volume":"21","author":"Fraenkel","year":"2014","journal-title":"Endocr.-Relat. Cancer"},{"key":"ref_7","first-page":"1","article-title":"A Robot Calibration Method Based on Joint Angle Division and an Artificial Neural Network","volume":"9","author":"Wang","year":"2019","journal-title":"Math. Probl. Eng."},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"105447","DOI":"10.1109\/ACCESS.2020.2999927","article-title":"Robot Manipulator Calibration Using a Model Based Identification Technique and a Neural Network with the Teaching Learning-Based Optimization","volume":"8","author":"Le","year":"2020","journal-title":"IEEE Access"},{"key":"ref_9","doi-asserted-by":"crossref","unstructured":"Shah, S.H., Lin, C.-Y., Tran, C.-C., and Ahmad, A.R. (2023). Robot Pose Estimation and Normal Trajectory Generation on Curved Surface Using an Enhanced Non-Contact Approach. Sensors, 23.","DOI":"10.3390\/s23083816"},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"150","DOI":"10.2478\/acss-2021-0018","article-title":"Time Series Forecasting of Mobile Robot Motion Sensors Using LSTM Networks","volume":"26","author":"Vagale","year":"2021","journal-title":"Appl. Comput. Syst."},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"3145","DOI":"10.1016\/j.procs.2025.04.572","article-title":"A Bi-LSTM based approach for Compensation detection during robotic stroke rehabilitation therapy","volume":"258","author":"Rani","year":"2025","journal-title":"Procedia Comput. Sci."},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"102629","DOI":"10.1016\/j.rcim.2023.102629","article-title":"BN-LSTM-based energy consumption modeling approach for an industrial robot manipulator","volume":"85","author":"Lin","year":"2023","journal-title":"Robot. Comput. Manuf."},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"138102","DOI":"10.1109\/ACCESS.2020.3012196","article-title":"Deep Learning Aided Dynamic Parameter Identification of 6-DOF Robot Manipulators","volume":"8","author":"Wang","year":"2020","journal-title":"IEEE Access"},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"137","DOI":"10.1016\/j.jmsy.2023.07.009","article-title":"Energy consumption prediction and optimization of industrial robots based on LSTM","volume":"70","author":"Jiang","year":"2023","journal-title":"J. Manuf. Syst."},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"173989","DOI":"10.1109\/ACCESS.2019.2957019","article-title":"Modeling and Simulation of Robot Inverse Dynamics Using LSTM-Based Deep Learning Algorithm for Smart Cities and Factories","volume":"7","author":"Liu","year":"2019","journal-title":"IEEE Access"},{"key":"ref_16","doi-asserted-by":"crossref","unstructured":"Rueckert, E., Nakatenus, M., Tosatto, S., and Peters, J. (2017, January 15\u201317). Learning inverse dynamics models in O(n) time with LSTM networks. Proceedings of the 2017 IEEE-RAS 17th International Conference on Humanoid Robotics (Humanoids), Birmingham, UK.","DOI":"10.1109\/HUMANOIDS.2017.8246965"},{"key":"ref_17","unstructured":"Reuss, M., van Duijkeren, N., Krug, R., Becker, P., Shaj, V., and Neumann, G. (2016). End-to-End Learning of Hybrid Inverse Dynamics Models for Precise and Compliant Impedance Control. arXiv."},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"4847","DOI":"10.1007\/s12206-023-0837-3","article-title":"LSTM-based external torque prediction for 6-DOF robot collision detection","volume":"37","author":"Zhang","year":"2023","journal-title":"J. Mech. Sci. Technol."},{"key":"ref_19","doi-asserted-by":"crossref","unstructured":"Huang, W., Lin, Y., Liu, M., and Min, H. (2024). Velocity-aware spatial-temporal attention LSTM model for inverse dynamic model learning of manipulators. Front. Neurorobotics, 18.","DOI":"10.3389\/fnbot.2024.1353879"},{"key":"ref_20","doi-asserted-by":"crossref","unstructured":"Kabir, R., Bhadra, D., Ridoy, M., and Milanova, M. (2025). LSTM\u2013Transformer-Based Robust Hybrid Deep Learning Model for Financial Time Series Forecasting. Sci, 7.","DOI":"10.3390\/sci7010007"},{"key":"ref_21","doi-asserted-by":"crossref","first-page":"33440","DOI":"10.1109\/ACCESS.2022.3161438","article-title":"Stochastically Forced Ensemble Dynamic Mode Decomposition for Forecasting and Analysis of Near-Periodic Systems","volume":"10","author":"Dylewsky","year":"2022","journal-title":"IEEE Access"},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"3","DOI":"10.1177\/027836498500400201","article-title":"Manipulability of Robotic Mechanisms","volume":"4","author":"Yoshikawa","year":"1985","journal-title":"Int. J. Robot. Res."},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"1061","DOI":"10.1243\/09544062JMES830","article-title":"Dimensional synthesis and dynamic manipulability of a planar two-degree-of-freedom parallel manipulator","volume":"222","author":"Wu","year":"2008","journal-title":"Proc. Inst. Mech. Eng. Part C J. Mech. Eng. Sci."}],"container-title":["Sensors"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1424-8220\/26\/2\/470\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,1,12]],"date-time":"2026-01-12T08:21:59Z","timestamp":1768206119000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1424-8220\/26\/2\/470"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,1,10]]},"references-count":23,"journal-issue":{"issue":"2","published-online":{"date-parts":[[2026,1]]}},"alternative-id":["s26020470"],"URL":"https:\/\/doi.org\/10.3390\/s26020470","relation":{},"ISSN":["1424-8220"],"issn-type":[{"value":"1424-8220","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,1,10]]}}}