VRBT：轻量级骨骼识别与损伤预警VR羽毛球训练方法

doi:10.16182/j.issn1004731x.joss.25-0862

系统仿真学报 ›› 2026, Vol. 38 ›› Issue (1): 225-234.doi: 10.16182/j.issn1004731x.joss.25-0862

• 论文 • 上一篇

VRBT：轻量级骨骼识别与损伤预警VR羽毛球训练方法

朱羽宁¹, 杨猛¹, 陈天玥¹, 孟维亮²^,³

^1.北京林业大学信息学院，北京 100083
^2.中国科学院自动化研究所多模态人工智能系统全国重点实验室，北京 100190
^3.中国科学院大学人工智能学院，北京 100049

收稿日期:2025-09-06 修回日期:2025-10-22 出版日期:2026-01-18 发布日期:2026-01-28
通讯作者: 杨猛
第一作者简介:朱羽宁(2004-)，女，本科生，研究方向为虚拟现实、人工智能等。
基金资助:
国家自然科学基金(62376271);国家自然科学基金(62572059);北京市自然科学基金(JQ23014)

VRBT: VR Badminton Training with Multitask Injury Alerts based on Lightweight 3D Skeletal Reconstruction

Zhu Yuning¹, Yang Meng¹, Chen Tianyue¹, Meng Weiliang²^,³

^1.School of Information Science and Technology, Beijing Forestry University, Beijing 100083, China
^2.State Key Laboratory of Multimodal Artificial Intelligence Systems, Institute of Automation, Chinese Academy of Sciences, Beijing 100190, China
^3.School of Artificial Intelligence, University of Chinese Academy of Sciences, Beijing 100049, China

Received:2025-09-06 Revised:2025-10-22 Online:2026-01-18 Published:2026-01-28
Contact: Yang Meng

摘要/Abstract

摘要：

针对传统羽毛球训练的场地与指导限制，提出一种基于多模型集成与协同仿真的VR羽毛球训练方法。构建了“感知-决策-交互”的闭环仿真框架：基于Unity引擎开发训练模拟、智能对抗及休闲游戏三大多场景训练模块，通过物理引擎实现运动轨迹仿真；采用轻量化MHFormer架构优化三维骨骼重建；设计多任务增强型损伤预测方法(enhanced injury prediction system, EIPS)，集成随机森林与XGBoost实现损伤概率、风险等级及预警类型的联合输出。通过上述模型的协同仿真，系统实现了对复杂人机交互过程的综合模拟与实时评估。所提方法可为解决骨骼重建实时性与精度的平衡难题提供思路，构建动态风险评分模型支撑个性化防护。

关键词: 虚拟现实, 羽毛球训练, 三维骨骼识别, 运动损伤预警, 多任务学习

Abstract:

To overcome the limitations of traditional badminton training, a VR training method that integrates multiple models for collaborative simulation is proposed. A "perception-decision- interaction" framework is developed within Unity, featuring diverse training modules powered by a physics engine for realistic trajectory simulation. The system employs a lightweight MHFormer for 3D pose estimation and a novel multi-task model (enhanced injury prediction system, EIPS) that combines random forest and XGBoost to jointly assess injury risk. This approach offers a solution for balancing real-time performance with accuracy in skeleton reconstruction and enables personalized training through dynamic risk assessment.

Key words: virtual reality(VR), badminton training, 3D skeleton recognition, sports injury early warning, multi-task learning

中图分类号:

TP391.9

朱羽宁,杨猛,陈天玥等 . VRBT：轻量级骨骼识别与损伤预警VR羽毛球训练方法[J]. 系统仿真学报, 2026, 38(1): 225-234.

Zhu Yuning,Yang Meng,Chen Tianyue,et al . VRBT: VR Badminton Training with Multitask Injury Alerts based on Lightweight 3D Skeletal Reconstruction[J]. Journal of System Simulation, 2026, 38(1): 225-234.

图/表 6

图1

表1

羽毛球在空中受到的力

力的类型	方向	计算公式	参数说明
重力	竖直向下	$F g = m ⋅ g$	m：球质量 g：重力加速度
空气阻力	与运动方向相反	$F d = 12 ρ v 2 C d A$	ρ：空气密度 v：瞬时速度 C_d：阻力系数 A：迎风面积
马格努斯力	垂直于运动与旋转轴	$F m = 12 ρ v 2 C L A$	C_L：升力系数依赖旋转角速度，旋转方向决定力的指向
浮力	竖直向上	$F b = ρ a i r V g$	V：羽毛球体积
侧风力	水平方向	$F w = 12 ρ v w 2 C w A$	v_w：风速 C_w：侧向阻力系数
粘性阻力	与运动方向相反	$F v = C v ⋅ A ⋅ μ ⋅ v L$	C_v：形状相关常数 A：物体表面积 L：特征长度

表1

图2

图3

表2

表3

参考文献 27

[1]	Neumann D L, Moffitt R L, Thomas P R, et al. A Systematic Review of the Application of Interactive Virtual Reality to Sport[J]. Virtual Reality, 2018, 22(3): 183-198.
[2]	Wang Jinzhao. Research on Application of Virtual Reality Technology in Competitive Sports[J]. Procedia Engineering, 2012, 29: 3659-3662.
[3]	Niu Yunfei, Dong Yan, Zhang Jianwei, et al. The Value of Athletics Class with the Use of Virtual Reality Games for Better Communication and Socialization of Students[J]. Education and Information Technologies, 2025, 30(1): 693-714.
[4]	Lee H Y, Chang C W, Chung C Y. Virtual Reality-Based Badminton Teaching in Physical Education Courses[J]. Physical Education Journal, 2020, 53(4): 375-392.
[5]	'Skip' Rizzo A, Shilling R. Clinical Virtual Reality Tools to Advance the Prevention, Assessment, and Treatment of PTSD[J]. European Journal of Psychotraumatology, 2017, 8(S5): 1414560.
[6]	Anderson-Hanley C, Snyder A L, Nimon J P, et al. Social Facilitation in Virtual Reality-enhanced Exercise: Competitiveness Moderates Exercise Effort of Older Adults[J]. Clinical Interventions in Aging, 2011, 6: 275-280.
[7]	Tai Weishen, Liu K H. Badminton Self-training System Based on Virtual Reality[C]//2023 Asia Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC). Piscataway: IEEE, 2023: 1659-1663.
[8]	Al Ardha M A, Nurhasan N, Wiriawan O, et al. Analysis of Badminton Research Trend in the Last Three Decades: Bibliographic Analysis of Scopus Journal Database[J]. Retos, 2024, 60: 129-139.
[9]	Gazali N, Cendra R, Efendi Y, et al. A New Dimension in Learning Badminton Refereeing: A 3D Simulation Android app with Virtual Reality Integration[J]. Jurnal Keolahragaan, 2024, 12(1): 29-39.
[10]	Li Yinghao, Jawis M. Construction and Effect of Interactive Teaching Methods in the College Badminton Classroom Using Fuzzy-assisted Virtual Reality Modeling: A Case Study in Chinese Universities[J]. Journal of Fuzzy Extension & Applications, 2024, 5(2): 199-222.
[11]	Ochi D, Kameda A, Takahashi K, et al. VR Technologies for Rich Sports Experience[C]//ACM SIGGRAPH 2016 Emerging Technologies. New York: ACM, 2016: 21.
[12]	Mushhood Afsar M, Saqib S. Virtual Reality Game Based Interaction with Exergaming Using Wearable Sensors[C]//2022 19th International Bhurban Conference on Applied Sciences and Technology (IBCAST). Piscataway: IEEE, 2022: 201-206.
[13]	Ye Shuainan, Chen Zhutian, Chu Xiangtong, et al. ShuttleSpace: Exploring and Analyzing Movement Trajectory in Immersive Visualization[J]. IEEE Transactions on Visualization and Computer Graphics, 2021, 27(2): 860-869.
[14]	Ginja G A. Applications of Virtual Reality in the Practice of Para-Badminton[C]//2018 20th Symposium on Virtual and Augmented Reality (SVR). Piscataway: IEEE, 2018: 230-232.
[15]	Liu Jiatong, Liang Bo. An Action Recognition Technology for Badminton Players Using Deep Learning[J]. Mobile Information Systems, 2022, 2022(1): 3413584.
[16]	Ren Bin, Liu Mengyuan, Ding Runwei, et al. A Survey on 3D Skeleton-based Action Recognition Using Learning Method[J]. Cyborg and Bionic Systems, 2024, 5: 0100.
[17]	Liu Yang, Liu Lei. Research on Openpose Method Based on Open Pose Bone RecognitionC]//2024 10th International Conference on Systems and Informatics (ICSAI). Piscataway: IEEE, 2024: 1-4.
[18]	Cai Cuiping. Badminton Video Analysis Based on Player Tracking and Pose Trajectory Estimation[C]//2021 13th International Conference on Measuring Technology and Mechatronics Automation (ICMTMA). Piscataway: IEEE, 2021: 471-474.
[19]	Zhou Bo, Yao Peng, Qiu Kaiyi, et al. Intelligent System of Badminton Serve Action Based on YOLOv5 and OpenPose[C]//2023 4th International Conference on Computer Engineering and Application (ICCEA). Piscataway: IEEE, 2023: 341-346.
[20]	Li Qi, Chiu T C, Huang H W, et al. VideoBadminton: A Video Dataset for Badminton Action Recognition[C]//2024 IEEE International Conference on Big Data (BigData). Piscataway: IEEE, 2024: 1387-1392.
[21]	Musat C L, Mereuta C, Nechita A, et al. Diagnostic Applications of AI in Sports: A Comprehensive Review of Injury Risk Prediction Methods[J]. Diagnostics, 2024, 14(22): 2516.
[22]	Xiang Chengqi. Early Warning Model of Track and Field Sports Injury Based on RBF Neural Network Algorithm[J]. Journal of Physics: Conference Series, 2021, 2037(1): 012084.
[23]	Hu Yi, Liu Jing. Real-time Monitoring and Early Warning System of Athlete Health Condition Based on Computer Internet of Things[C]//2024 IEEE 2nd International Conference on Sensors, Electronics and Computer Engineering (ICSECE). Piscataway: IEEE, 2024: 1392-1398.
[24]	Yang Gui, XU Xiao. Sequence Video and Artificial Intelligence Assisted Basketball Injury Risk Early Warning Method[J]. Mobile Information Systems, 2022, 2022(1): 4778245.
[25]	Mroz S, Baddour N, Mcguirk C, et al. Comparing the Quality of Human Pose Estimation with BlazePose or OpenPose[C]//2021 4th International Conference on Bio-engineering for Smart Technologies (BioSMART). Piscataway: IEEE, 2021: 1-4.
[26]	Bazarevsky V, Grishchenko I, Raveendran K, et al. BlazePose: On-device Real-time Body Pose Tracking[EB/OL]. (2020-06-17) [2025-07-16]. .
[27]	Li Wenhao, Liu Hong, Tang Hao, et al. MHFormer: Multi-hypothesis Transformer for 3D Human Pose Estimation[EB/OL]. (2022-06-28) [2025-07-16]. .

主要参数	量值
损伤概率判别能力 (AUC)	0.953 ± 0.012
风险等级分类准确率(Accuracy)	0.945 ± 0.015
预警类型识别性能 (Micro-F1 Score)	0.935 ± 0.018

模型	AUC	Accuracy	Micro-F1 Score
XGBoost	0.530	0.214	0.961
Random Forest	0.897	0.723	0.703
GBDT	0.929	0.836	0.824
DART&GBDT	0.932	0.834	0.820
本文模型	0.953	0.945	0.935

VRBT：轻量级骨骼识别与损伤预警VR羽毛球训练方法

VRBT: VR Badminton Training with Multitask Injury Alerts based on Lightweight 3D Skeletal Reconstruction

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

图/表 6

参考文献 27

相关文章 15

编辑推荐

Metrics

本文评价

[1]	曲静, 方凯宁, 朱山童, 卜令国. 基于多模态脑机接口的虚拟现实康复训练系统[J]. 系统仿真学报, 2026, 38(1): 125-135.
[2]	陈真, 吴卓屹, 张霖. 深度强化学习中策略表征研究简述[J]. 系统仿真学报, 2025, 37(7): 1753-1769.
[3]	龚建兴, 胡海, 任海慧, 吴瑞祥. 面向虚实结合的军事训练系统互操作模型与运用[J]. 系统仿真学报, 2025, 37(12): 3161-3175.
[4]	王子宁, 罗诗江, 陈维亚, 骆汉宾. 沉浸式钢架雪车滑行仿真系统研究与实现[J]. 系统仿真学报, 2025, 37(1): 145-154.
[5]	王红军, 林俊强, 邹湘军, 张坡, 周铭轩, 邹伟锐, 唐昀超, 罗陆锋. 基于数字孪生的果园虚拟交互系统构建[J]. 系统仿真学报, 2024, 36(6): 1493-1508.
[6]	张昭, 郭玉杰, 赵晓宁, 孙宝亮, 邓双厚, 冯国旭. 军事元宇宙刍议与展望[J]. 系统仿真学报, 2023, 35(7): 1421-1437.
[7]	姚舜, 胡忠志, 曹文宇, 杨佳利. 沉浸式航空发动机视景系统的设计与实现[J]. 系统仿真学报, 2023, 35(6): 1395-1404.
[8]	权巍, 王超, 耿雪娜, 韩成. 基于运动感知的VR体验舒适度研究[J]. 系统仿真学报, 2023, 35(1): 169-177.
[9]	周晓光, 朱鹏, 张源原, 路欢, 周媛. 六自由度跳伞模拟训练系统设计与实现[J]. 系统仿真学报, 2022, 34(6): 1320-1329.
[10]	许世健, 赵丹, 苏铖宇, 王赋攀, 张晓蓉, 王昉, 吴亚东. 面向流场可视化的沉浸式虚拟现实交互系统研究[J]. 系统仿真学报, 2022, 34(5): 1160-1172.
[11]	彭博, 汪强, 青芮冰, 尹立雪, 姜劲枫. 基于光线追踪的实时超声模拟与虚拟现实的集成[J]. 系统仿真学报, 2022, 34(11): 2425-2436.
[12]	邵绪强, 张浩伟, 冯小华. 面向电力变压器虚拟装配的多感官融合交互方法[J]. 系统仿真学报, 2022, 34(10): 2244-2254.
[13]	孙佳文, 任鸿翔, 肖方兵, 蒋效彬. 基于虚拟现实的整船仿真训练平台[J]. 系统仿真学报, 2021, 33(9): 2243-2251.
[14]	魏士松, 周正东, 章栩苓, 毛玲, 贾峻山, 刘传乐. 基于桌面虚拟现实技术的航天器虚拟维修训练系统[J]. 系统仿真学报, 2021, 33(6): 1358-1363.
[15]	贾春洋, 邹湘军, 李锦慧, 曾泽钦, 王杰. 虚拟环境下的复杂变速箱交互式装配仿真平台[J]. 系统仿真学报, 2020, 32(9): 1736-1743.