An explainable AI framework for badminton stroke quality assessment via hierarchical fuzzy inference and LLM-assisted coaching analytics

i

This article has been updated. A Correction to this article was published on .

i

The Original Article was published on

i

The Original Article was published on

i

This article has been updated

i

This article was retracted on

i

This article has been updated

Abstract

This study proposes an explainable computational intelligence framework for badminton stroke-quality assessment with large language model (LLM)-assisted coaching analytics. The proposed framework integrates stroke event segmentation, skeletal motion analysis, graph-based action recognition, stroke-quality assessment, and natural-language feedback generation. Racket–shuttlecock contact moments are first detected to extract individual stroke clips. MediaPipe Pose is employed to obtain skeletal joint trajectories, from which six stroke performance indices are formulated, including swing velocity index, wrist trajectory smoothness index, wrist posture index, elbow posture index, footwork efficiency index, and racket control index. These indices are integrated to generate interpretable stroke-quality ratings and radar-chart visualizations. For stroke classification, the proposed mST-GCN++ model achieved 94.34% accuracy across 11 badminton stroke categories, while the racket–shuttlecock contact detection module achieved an F1-score of 99.74%. The recognized stroke type, stroke performance indices, and assessment results are subsequently provided to an LLM-based module to generate personalized coaching feedback. Experimental results demonstrate that the proposed framework can effectively distinguish effective and ineffective stroke executions through interpretable quantitative assessments. Human-in-the-loop evaluation with badminton experts and players achieved an average score of 8.594/10, indicating that the generated feedback is technically accurate, consistent with observed movements, and useful for practical training. The proposed framework contributes by introducing six interpretable stroke performance indices that bridge low-level skeletal motion and high-level coaching knowledge, a hierarchical fuzzy inference system that performs transparent biomechanical reasoning for stroke-quality assessment, and a knowledge-guided LLM-assisted coaching framework that generates grounded natural-language coaching feedback from structured assessment results.

Keywords

badminton stroke assessment / fuzzy inference system / human pose estimation / large language model / skeleton-based action recognition / spatio-temporal graph convolutional networks

CHANGE HISTORY

A Correction to this paper has been published:
https://doi.org

The Retraction Note was updated to reflect reception of statements of disagreement from all authors.

This article has been retracted.Please see the Retraction Notice for more detail:
https://doi.org

References

[1] G. Bisht, V. B. Mishra, Biomechanical performance assessment of overhead backhand smash strokes of male badminton players, Int. J. Phys. Educ. Sports Health, 12 (2025), 159-161. https://doi.org/10.22271/kheljournal.2025.v12.i3c.3791

[2] H. Huang, Z. Guo, M. Zhao, M. Liu, J. Dai, Differences in backcourt forehand clear stroke between novice players and experienced badminton players: Based on body segment acceleration data, BMC Sports Sci. Med. Rehabil., 17 (2025), 118. https://doi.org/10.1186/s13102-025-01163-w

[3] X. Cui, W. K. Lam, Q. Gao, X. Wang, T. Zhao, Biomechanical analysis on skilled badminton players during take-off phase in forehand overhead strokes: A pilot study, Biology, 11 (2022), 1401. https://doi.org/10.3390/biology11101401

[4] F. Li, S. Li, X. Zhang, G. Shan, Biomechanical insights for developing evidence-based training programs: Unveiling the kinematic secrets of the overhead forehand smash in badminton through novice-skilled player comparison, Appl. Sci. (Basel), 13 (2023), 12488. https://doi.org/10.3390/app132212488

[5] E. Edmizal, E. Barlian, D. Donie, A. Komaini, T. H. Sin, U. Umar, et al., Biomechanical analysis of smash stroke in badminton: A comparative study of elite and recreational players: A systematic review, Retos, 57 (2024), 809-817.

[6] I. Ghosh, S. R. Ramamurthy, A. Chakma, N. Roy, DeCoach: Deep learning-based coaching for badminton player assessment, Pervasive Mob. Comput., 83 (2022), 101608. https://doi.org/10.1016/j.pmcj.2022.101608

[7] X. Zhu, L. Liu, J. Huang, G. Chen, X. Ling, Y. Chen, The analysis of motion recognition model for badminton player movements using machine learning, Sci. Rep., 15 (2025), 19030. https://doi.org/10.1038/s41598-025-02771-9

[8] Y. Wang, L. Wu, Design of badminton technical movement recognition system based on improved Agnes algorithm, J. Electr. Syst., 20 (2024), 1981-1991. https://doi.org/10.52783/jes.3113

[9] Y. Jia, N. A. Abdullah, H. Eliza, Q. Lu, D. Si, H. Guo, et al., A narrative review of deep learning applications in sports performance analysis: Current practices, challenges, and future directions, BMC Sports Sci. Med. Rehabil., 17 (2025), 249. https://doi.org/10.1186/s13102-025-01294-0

[10] L. Xiao, Y. Cao, Y. Gai, E. Khezri, J. Liu, M. Yang, Recognizing sports activities from video frames using deformable convolution and adaptive multiscale features, J. Cloud Comput., 12 (2023), 167. https://doi.org/10.1186/s13677-023-00552-1

[11] Y. H. Hsu, C. C. Yu, H. Y. Cheng, Enhancing badminton game analysis: An approach to shot refinement via a fusion of shuttlecock tracking and hit detection from monocular camera, Sensors, 24 (2024), 4372. https://doi.org/10.3390/s24134372

[12] P. Liu, J. Wang, MonoTrack: Shuttle trajectory reconstruction from monocular badminton video, 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, New Orleans, LA, USA, (2022), 3512-3521. https://doi.org/10.1109/CVPRW56347.2022.00395

[13] J. Y. Chang, BST: Badminton stroke-type transformer for skeleton-based action recognition in racket sports, preprint, arXiv:2502.21085.

[14] Y. Lai, Z. Shi, C. Zhu, YO-CSA-T: A real-time badminton tracking system utilizing YOLO based on contextual and spatial attention, in 2025 IEEE/RSJ International Conference on Intelligent Robots and Systems, Hangzhou, China, (2025), 732-739. https://doi.org/10.1109/IROS60139.2025.11246442

[15] Z. Cao, T. Liao, W. Song, Z. Chen, C. Li, Detecting the shuttlecock for a badminton robot: A YOLO based approach, Expert Syst. Appl., 164 (2021), 113833. https://doi.org/10.1016/j.eswa.2020.113833

[16] Y. C. Huang, I. N. Liao, C. H. Chen, T. U. İk, W. C. Peng, TrackNet: A deep learning network for tracking high-speed and tiny objects in sports applications, in 2019 16th IEEE International Conference on Advanced Video and Signal Based Surveillance , Taipei, Taiwan, (2019), 1-8. https://doi.org/10.1109/AVSS.2019.8909871

[17] Y. J. Chen, Y. S. Wang, TrackNetV3: Enhancing shuttlecock tracking with augmentations and trajectory rectification, in 5th ACM International Conference on Multimedia in Asia, Tainan, Taiwan, (2023), 1-7. https://doi.org/10.1145/3595916.3626377

[18] Y. Wu, Z. Chen, H. Zhang, Y. Yang, W. Yi, Enhanced pose estimation for badminton players via improved YOLOv8-pose with efficient local attention, Sensors (Basel), 25 (2025), 4446. https://doi.org/10.3390/s25144446

[19] Q. Zhang, L. Cheng, Research on deep learning based badminton players’ movement analysis methods, in 2024 2nd International Conference on Signal Processing and Intelligent Computing , Guangzhou, China, (2024), 1175-1178. https://doi.org/10.1109/SPIC62469.2024.10691578

[20] P. Harris, G. S. Mathi, S. Sivadarsini, B. T. Jenifer, Methods of postural assessment in badminton: A comparative analysis of conventional techniques and AI-powered pose estimation, Int. J. Adv. Res. Comput. Sci., 16 (2025), 49. https://doi.org/10.26483/ijarcs.v16i3.7245

[21] D. Rode, A. Dunkel, R. Willi, P. Wolf, M. Xiloyannis, R. Riener, Assessment of monocular human pose estimation models for clinical movement analysis, Sci. Rep., 15 (2025), 38767. https://doi.org/10.1038/s41598-025-22626-7

[22] G. H. Martínez, Openpose: Whole-body Pose Estimation , Ph.D thesis, Carnegie Mellon University in Pittsburgh, PA, USA, 2019.

[23] J. W. Kim, J. Y. Choi, E. J. Ha, J. H. Choi, Human pose estimation using mediapipe pose and optimization method based on a humanoid model, Appl. Sci., 13 (2023), 2700. https://doi.org/10.3390/app13042700

[24] V. Bazarevsky, I. Grishchenko, K. Raveendran, T. Zhu, F. Zhang, M. Grundmann, BlazePose: On-device real-time body pose tracking, preprint, arXiv:2006.10204.

[25] A. Kini, S. Rubini, PoseNet: A novel YOLO-driven framework for badminton posture detection and correction, in International Conference on ICT for Sustainable Development, Goa, India, (2025), 227-239. https://doi.org/10.1007/978-3-032-06662-6_22

[26] W. Liu, Q. Bao, Y. Sun, T. Mei, Recent advances of monocular 2D and 3D human pose estimation: A deep learning perspective, ACM Comput. Surv., 55 (2022), 1-41. https://doi.org/10.1145/3524497

[27] S. Yan, Y. Xiong, D. Lin, Spatial temporal graph convolutional networks for skeleton-based action recognition, Proc. AAAI Conf. Artif. Intell., 32 (2018), 7444-7452. https://doi.org/10.1609/aaai.v32i1.12328

[28] L. Shi, Y. Zhang, J. Cheng, H. Lu, Skeleton-based action recognition with multi-stream adaptive graph convolutional networks, IEEE Trans. Image Process., 29 (2019), 9532-9545. https://doi.org/10.1109/tip.2020.3028207

[29] L. Yu, L. Tian, Q. Du, J. A. Bhutto, Multi-stream adaptive spatial-temporal attention graph convolutional network for skeleton-based action recognition, IET Comput. Vis., 16 (2021), 143-158. https://doi.org/10.1049/cvi2.12075

[30] U. Kilic, O. O. Karadag, G. T. Ozyer, AGMS-GCN: Attention-guided multi-scale graph convolutional networks for skeleton-based action recognition, Knowl. Based Syst., 311 (2025), 113045. https://doi.org/10.1016/j.knosys.2025.113045

[31] Y. Cao, C. Liu, Z. Huang, Y. Sheng, Y. Ju, Skeleton-based action recognition with temporal action graph and temporal adaptive graph convolution structure, Multimed. Tools Appl., 80 (2021), 29139-29162. https://doi.org/10.1007/s11042-021-11136-z

[32] Y. Chen, Z. Zhang, C. Yuan, B. Li, Y. Deng, W. Hu, Channel-wise topology refinement graph convolution for skeleton-based action recognition, preprint, arXiv:2107.12213.

[33] D. Chen, M. Chen, P. Wu, M. Wu, T. Zhang, C. Li, Two-stream spatio-temporal GCN-transformer networks for skeleton-based action recognition, Sci. Rep., 15 (2025), 4982. https://doi.org/10.1038/s41598-025-87752-8

[34] T. Yagneshwar, S. Mukherjee, STTGC-Net: Spatial-temporal transformer with graph convolution for skeleton-based action recognition, in 14th Indian Conference on Computer Vision, Graphics and Image Processing, New York, NY, USA, (2023), 1-10. https://doi.org/10.1145/3627631.3627640

[35] M. A. Sarwar, Y. C. Lin, Y. A. Daraghmi, T. U. İk, Y. L. Li, Skeleton based keyframe detection framework for sports action analysis: Badminton smash case, IEEE Access, 11 (2023), 90891-90900. https://doi.org/10.1109/access.2023.3307620

[36] F. Asriani, A. Azhari, W. Wahyono, Enhanced badminton stroke recognition using hybrid RGB-skeleton features and ensemble learning, Eng. Technol. Appl. Sci. Res., 16 (2026), 31303-31309. https://doi.org/10.48084/etasr.15586

[37] Y. Li, Y. Feng, X. Wang, G. Lu, Deep learning-based badminton action recognition and quality assessment, Intell. Data Anal., 30 (2025), 531-543. https://doi.org/10.1177/1088467x251353444

[38] R. Pang, X. Jiang, Research on image recognition technology to assess the quality of badminton players’ stroke action and training improvement strategy, Int. J. Comput. Inf. Syst. Ind. Manage. Appl., 18 (2026), 1-9. https://doi.org/10.70917/ijcisim-2026-0083

[39] T. Chen, K. Chen, X. Liu, P. Ke, Z. Sun, BadminSense: Enabling fine-grained badminton strokes evaluation on single smartwatch, in 2026 CHI Conference on Human Factors in Computing Systems, Barcelona, Spain, (2026), 1-20. https://doi.org/10.1145/3772318.3790998

[40] M. Seong, G. Kim, D. Yeo, Y. Kang, H. Yang, J. DelPreto, et al., Wearable sensor-based biomechanical dataset for evaluation of badminton performance, Sci. Data, 11 (2024), 343. https://doi.org/10.1038/s41597-024-03144-z

[41] X. Lai, Y. Lai, J. Chen, S. Huang, Q. Gao, C. Huang, Evaluation strategies for large language model-based models in exercise and health coaching: Scoping review, J. Med. Internet Res., 27 (2025), 79217. https://doi.org/10.2196/79217

[42] X. Lai, J. Chen, Y. Lai, S. Huang, Y. Cai, Z. Sun, et al., Using large language models to enhance exercise recommendations and physical activity in clinical and healthy populations: Scoping review, JMIR Med. Inf., 13 (2025), 59309. https://doi.org/10.2196/59309

[43] Z. He, J. Wang, B. Zhang, Y. Li, Knowledge-grounded large language model for personalized sports training plan generation, Sci. Rep., 16 (2026), 6793. https://doi.org/10.1038/s41598-026-37075-z

[44] N. Hegde, M. Vardhan, D. Nathani, E. Rosenzweig, C. Speed, A. Karthikesalingam, et al., Infusing behavior science into large language models for activity coaching, PLoS Digit. Health, 3 (2023), 0000431. https://doi.org/10.1371/journal.pdig.0000431

[45] Y. Wang, X. Feng, Y. Wu, X. Zhang, J. Wu, CoachXNet: An artificial intelligence and internet of things integrated platform for personalized training and feedback in digital sports, Int. J. Comput. Intell. Syst., 19 (2026), 49. https://doi.org/10.1007/s44196-025-01146-2

[46] O. Bodemer, Enhancing individual sports training through artificial intelligence: A comprehensive review, preprint, TechRxiv:24005916.v1.

[47] Y. Han, Intent-aware personalized feedback generation from coach-athlete dialogues in sports training, J. King Saud Univ. Comput. Inf. Sci., 37 (2025), 150. https://doi.org/10.1007/s44443-025-00165-5

[48] E. Weissblueth, A. Hirsh, I. Linder, Human expertise and artificial intelligence in coaching: A comparative analysis of feedback for triathlon front crawl technique, Int. J. Sports Sci. Coach, 0 (2026), 17479541261435125. https://doi.org/10.1177/17479541261435125

[49] J. Li, Y. Shen, W. Zhou, H. Chen, SearchExpert: A GenAI-driven framework for reasoning-intensive multimedia information fusion through fine-tuning and reinforcement learning, Inf. Fusion, 126 (2026), 103665. https://doi.org/10.1016/j.inffus.2025.103665

[50] X. Zhu, R. Wang, L. Cascone, E. Xiaosong, H. Chen, EUPA: Embodied uncertainty-aware multimodal affect perception for consumer electronic systems, IEEE Trans. Consum. Electron., 72 (2026), 6508-6515. https://doi.org/10.1109/TCE.2026.3679325

[51] X. Zhu, Z. Liu, E. Cambria, X. Yu, X. Fan, H. Chen, et al., A client-server based recognition system: Non-contact single/multiple emotional and behavioral state assessment methods, Comput. Methods Programs Biomed., 260 (2025), 108564. https://doi.org/10.1016/j.cmpb.2024.108564

[52] Q. Li, R. Li, K. Ji, W. Dai, Kalman filter and its application, in 2015 8th International Conference on Intelligent Networks and Intelligent Systems, Tianjin, China, (2015), 74-77. https://doi.org/10.1109/ICINIS.2015.35

[53] H. Duan, J. Wang, K. Chen, D. Li, PYSKL: Towards good practices for skeleton action recognition, in 30th ACM International Conference on Multimedia, Lisboa, Portugal, (2022), 7351-7354. https://doi.org/10.1145/3503161.3548545

[54] Y. A. Farha, J. Gall, MS-TCN: Multi-stage temporal convolutional network for action segmentation, in 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Long Beach, CA, USA, (2019), 3575-3584.

[55] Y. Nie, N. H. Nguyen, P. Sinthong, J. Kalagnanam, A time series is worth 64 words: Long-term forecasting with transformers, preprint, arXiv:2211.14730.

[56] H. J. Zimmermann, Fuzzy Set Theory and Its Applications , 4th edition, Springer Dordrecht, New York, 2011. https://doi.org/10.1007/978-94-010-0646-0

[57] R. Khanam, M. Hussain, YOLOv11: An overview of the key architectural enhancements, preprint, arXiv:2410.17725.