Research on Human-robot Collaborative Assembly Sequence Replanning Based on the Human Muscle Activation Fatigue Model

doi:10.3901/JME.2025.15.339

Abstract

Abstract: During human-robot collaborative assembly, prolonged repetitive work can lead to muscle fatigue in the human body, causing fluctuations in the completion time of assembly sequences and subsequently reducing overall assembly efficiency. Therefore, enhancing the robot’s ability to perceive the current fatigue state of the human, assessment of muscle fatigue during assembly in real-time and optimization of assembly sequences are of great significance for improving assembly efficiency and alleviating the human’s workload. To this end, this paper studies the replanning method for human-robot collaborative assembly sequences based on a muscle activation fatigue model. Firstly, motion images of the human during the assembly process are collected, and motion data are extracted using OpenPose to construct a personalized skeletal muscle model for the individual. Next, OpenSim is used to calculate the changes in muscle force during assembly actions, and the fatigue index of the most fatigued muscles executing the assembly actions is calculated in conjunction with the muscle activation fatigue model. Real-time joint angle variation data are utilized to train an LSTM deep neural network to quickly determine the action category of the human based on real-time joint angle, and then the fatigue index of the human’s real-time actions can be determined in real-time. Subsequently, a multi-objective genetic algorithm is employed to optimize the total assembly time and total human fatigue. Based on the real-time joint angle, the human action category and the corresponding fatigue index can be obtained. When the fatigue accumulates in the fatigued muscles, the assembly sequence in the next round is re-planned based on the real-time monitored human fatigue and action duration, thereby improving overall assembly efficiency and reducing human fatigue. The human-robot collaborative assembly of gear reducers is adopted as the experimental scenario, real-time fatigue detection and assembly sequence re-planning experiments are conducted. The experimental results indicate that the proposed method can identify muscle fatigue in humans and effectively re-plan assembly sequences that can alleviate fatigue, enhancing the efficiency of human-robot collaborative assembly and reducing muscle fatigue during the assembly.

Key words: human-robot collaborative assembly, human muscle activation fatigue model, sequence replanning, fatigue assessment, multi-objective genetic algorithm

CLC Number:

TP242

YAO Bitao, XIANG Yuguang, XU Wenjun, YE Xun, ZHOU Zude, WANG Lihui. Research on Human-robot Collaborative Assembly Sequence Replanning Based on the Human Muscle Activation Fatigue Model[J]. Journal of Mechanical Engineering, 2025, 61(15): 339-350.

References

[1] LENG J, SHA W, WANG B, et al. Industry 5.0:Prospect and retrospect[J]. Journal of Manufacturing Systems, 2022, 65:279-295.
[2] VYSOCKY A, NOVAK P. Human-robot collaboration in industry[J]. MM Science Journal, 2016, 2016(2):903-906.
[3] ABDOUS M A, DELORME X, BATTINI D, et al. Multi-objective optimization of assembly lines with workers fatigue consideration[J]. Ifac Papersonline, 2018, 51(11):698-703.
[4] MA L, CHABLAT D, BENNIS F, et al. A new muscle fatigue and recovery model and its ergonomics application in human simulation[J]. Virtual and Physical Prototyping, 2010, 5(3):123-137.
[5] MA L, CHABLAT D, BENNIS F, ZHANG W. A new simple dynamic muscle fatigue model and its validation[J]. International Journal of Industrial Ergonomics, 2009, 39(1):211-220.
[6] CHAND S, ZHENG H, LU Y. A vision-enabled fatiguesensitive human digital twin towards human-centric human-robot collaboration[J]. Journal of Manufacturing Systems, 2024, 77:432-445.
[7] YAACOUB A, THOMAS V, COLAS F, MAURICE P. A Probabilistic model for cobot decision making to mitigate human fatigue in repetitive co-manipulation tasks[J]. IEEE Robotics and Automation Letters, 2023, 8(11):7352-7359.
[8] YAO B, LI X, JI Z, et al. Task reallocation of human-robot collaborative production workshop based on a dynamic human fatigue model[J]. Computers&Industrial Engineering, 2024, 189:109855.
[9] YOU Y, CAI B, PHAM D T, et al. A human digital twin approach for fatigue-aware task planning in human-robot collaborative assembly[J]. Computers&Industrial Engineering, 2025, 200:11077.
[10] ZHENG H, CHAND S, KESHVARPARAST A, et al. Video-based fatigue estimation for human-robot task allocation optimisation[C]//2023 IEEE 19th International Conference on Automation Science and Engineering (CASE). IEEE, 2023:1-6.
[11] PETZOLDT C, NIERMANN D, MAACK E, et al. Implementation and evaluation of dynamic task allocation for human-robot collaboration in assembly[J]. Applied Sciences, 2022, 12(24):12645.
[12] KHEZRI A, BENDERBAL H H, BENYOUCEF L. Towards a sustainable reconfigurable manufacturing system (SRMS):Multi-objective based approaches for process plan generation problem[J]. International Journal of Production Research, 2020, 59(15):4533-4558.
[13] XU W, TANG Q, LIU J, et al. Disassembly sequence planning using discrete Bees algorithm for human-robot collaboration in remanufacturing[J]. Robotics and Computer-Integrated Manufacturing, 2020, 62:101860.
[14] YASSI A. Work-related musculoskeletal disorders[J]. Current Opinion in Rheumatology, 2000, 12(2):124-130.
[15] DELP S L, ANDERSON F C, ARNOLD A S, et al. OpenSim:Open-source software to create and analyze dynamic simulations of movement[J]. IEEE Trans Biomed Eng, 2007, 54(11):1940-1950.
[16] DEB K, PRATAP A, AGARWAL S, et al. A fast and elitist multiobjective genetic algorithm:NSGA-Ⅱ[J]. IEEE Transactions on Evolutionary Computation, 2002, 6(2):182-197.
[17] CAO Z, HIDALGO G, SIMON T, et al. OpenPose:Realtime multi-person 2D pose estimation using part affinity fields[J]. IEEE Trans Pattern Anal Mach Intell, 2021, 43(1):172-186.
[18] PAGNON D, DOMALAIN M, REVERET L. Pose2Sim:An open-source Python package for multiview markerless kinematics[J]. Journal of Open Source Software, 2022, 7(77):4362.
[19] JIANG T, LU P, ZHANG L, et al. Rtmpose:Real-time multi-person pose estimation based on mmpose[J]. arxiv preprint arxiv:2303.07399, 2023.
[20] MCFARLAND D C, MCCAIN E M, POPPO M N, et al. Spatial dependency of glenohumeral joint stability during dynamic unimanual and bimanual pushing and pulling[J]. Journal of Biomechanical Engineering, 2019, 141(5):051006.
[21] HOCHREITER S, SCHMIDHUBER J. Long short-term memory[J]. Neural Comput, 1997, 9(8):1735-1780.
[22] KINGMA D P, BA J. Adam:A method for stochastic optimization[J]. arxiv preprint arxiv:1412.6980, 2014.