This study investigated transverse pelvis rotations across walking, jogging, running, and sprinting using minimal limit cycle (MLC) modeling. Ten soccer players performed four gaits while a sacrum-mounted inertial measurement unit (IMU) recorded kinematics. We hypothesized that one expression could capture control mechanisms across gaits. A unified MLC model described all gait types within a common dynamical structure, with differences explained by coefficient modulation rather than distinct control strategies. Nonlinearity decreased from walking to sprinting, consistent with energetic differences between pendulum-like walking and spring–mass running dynamics. Stiffness coefficients (linear and nonlinear) varied significantly: linear stiffness was higher in sprinting than walking and jogging, while nonlinear stiffness was reduced in sprinting compared to jogging. These adaptations suggest gait-specific modulation of pelvis rotational velocity within an overarching shared framework, consistent with (but not directly demonstrating) central pattern generator (CPG)-based control architecture. Damping coefficients showed less sensitivity across gaits. Nonlinear stiffness (Duffing) and damping (Van der Pol) coefficients were consistent with stabilization mechanisms related to momentum preservation and energy recycling. Model fit improved from ∼63% in walking to ∼93% in sprinting, without changing equation structure, highlighting parameter modulation as the key mechanism. Quantifying gait transitions with a single pelvis sensor supports field-based monitoring, rehabilitation, and movement-quality screening related to injury risk, while informing bipedal robotics and rehabilitation devices requiring unified control architectures. To our knowledge, this is the first study to report unified gait modeling across four gait types using MLC with a single sensor, underscoring translational value in biomechanics, clinical practice, and robotics.
MastersSEChallisJH. Increasing the stability of the spring-loaded inverted pendulum model of running with a wobbling mass. J Biomech2021; 123: 110527. https://doi.org/10.1016/j.jbiomech.2021.110527
3.
LuLTanYKlaicM, et al. Evaluating rehabilitation progress using motion features identified by machine learning. IEEE Trans Biomed Eng2021; 68(4): 1417–1428. https://doi.org/10.1109/TBME.2020.3036095
LeckeyCvan DykNDohertyC, et al. Machine learning approaches to injury risk prediction in sport: a scoping review with evidence synthesis. Br J Sports Med2025; 59: 491–500. https://doi.org/10.1136/bjsports-2024-108576
6.
DomaradzkiJKopackaE. Predictive models for injury risk across body regions and sport types in physically active students: cross-sectional design. J Clin Med2025; 14(12): 4307. https://doi.org/10.3390/jcm14124307
BaoYYangHW. A three-dimensional spring-loaded inverted pendulum walking model considering human movement speed and frequency. Bioinspir Biomim2024; 19(4): 046012. https://doi.org/10.1088/1748-3190/ad48ee
9.
AbdullahMHulleckAAKatmahR, et al. Multibody dynamics-based musculoskeletal modeling for gait analysis: a systematic review. J Neuroeng Rehabil2024; 21: 178. https://doi.org/10.1186/s12984-024-01458-y
10.
NitschkeMDorschkyELeyendeckerS, et al. Estimating 3D kinematics and kinetics from virtual inertial sensor data through musculoskeletal movement simulations. Front Bioeng Biotechnol2024; 12: 1285845. https://doi.org/10.3389/fbioe.2024.1285845
ZehrEPBalterJEFerrisDP, et al. Neural regulation of rhythmic arm and leg movement is conserved across human locomotor tasks. J Physiol2007; 582(Pt 1): 209–227. https://doi.org/10.1113/jphysiol.2007.127237
14.
DiedrichFJWarrenWHJr. Why change gaits? Dynamics of the walk-run transition. J Exp Psychol Hum Percept Perform1995; 21(1): 183–202. https://doi.org/10.1037/0096-1523.21.1.183
CappelliniGIvanenkoYPPoppeleRE, et al. Motor patterns in human walking and running. J Neurophysiol2006; 95(6): 3426–3437. https://doi.org/10.1152/jn.00081.2006
17.
FioriLCastigliaSFChiniG, et al. The lower limb muscle co-activation map during human locomotion: from slow walking to running. Bioengineering2024; 11(3): 288. https://doi.org/10.3390/bioengineering11030288
18.
DarendeliAErtanHEnokaRM. Comparison of EMG activity in leg muscles between overground and treadmill running. Med Sci Sports Exerc2023; 55(3): 517–524. https://doi.org/10.1249/MSS.0000000000003055
LamothCJDaffertshoferAHuysR, et al. Steady and transient coordination structures of walking and running. Hum Mov Sci2009; 28(3): 371–386. https://doi.org/10.1016/j.humov.2008.10.001
21.
MacKay-LyonsM. Central pattern generation of locomotion: a review of the evidence. Phys Ther2002; 82(1): 69–83. https://doi.org/10.1093/ptj/82.1.69
22.
MinassianKBayartALacknerP, et al. Rare phenomena of central rhythm and pattern generation in a case of complete spinal cord injury. Nat Commun2023; 14: 3276. https://doi.org/10.1038/s41467-023-39034-y
23.
SuHJingLLvD, et al. Central pattern generators in spinal cord injury: mechanisms, modulation, and therapeutic strategies for motor recovery. JOR Spine2025; 8(3): e70100. https://doi.org/10.1002/jsp2.70100
24.
McAndrewPMDingwellJBWilkenJM. Walking variability during continuous pseudo-random oscillations of the support surface and visual field. J Biomech2010; 43(8): 1470–1475. https://doi.org/10.1016/j.jbiomech.2010.02.003
25.
O’ConnorSMKuoAD. Direction-dependent control of balance during walking and standing. J Neurophysiol2009; 102(3): 1411–1419. https://doi.org/10.1152/jn.00131.2009
26.
McAndrewPMWilkenJMDingwellJB. Dynamic stability of human walking in visually and mechanically destabilizing environments. J Biomech2011; 44(4): 644–649. https://doi.org/10.1016/j.jbiomech.2010.11.007
27.
WurdemanSRStergiouN. Temporal structure of variability reveals similar control mechanisms during lateral stepping and forward walking. Gait Posture2013; 38(1): 73–78. https://doi.org/10.1016/j.gaitpost.2012.10.017
28.
MehdizadehSArshiARDavidsK. Constraints on dynamic stability during forward, backward and lateral locomotion in skilled football players. Eur J Sport Sci2016; 16(2): 190–198. https://doi.org/10.1080/17461391.2014.995233
29.
AlavikiaMArshiAR. Enhancing assessment of intensive change of direction: minimal models for pelvis movements during approach and exit phases. Comput Methods Biomech Biomed Engin2025: 1–14. https://doi.org/10.1080/10255842.2025.2578702
30.
McGinnisRSCainSMDavidsonSP, et al. Inertial sensor and cluster analysis for discriminating agility run technique and quantifying changes across load. J Biomech Eng2016; 138(12): 124503. https://doi.org/10.1115/1.4034717
31.
TrecrociARossiADos’SantosT, et al. Change of direction asymmetry across different age categories in youth soccer. PeerJ2020; 8: e9486. https://doi.org/10.7717/peerj.9486
32.
HewettTEMyerGDFordKR, et al. Mechanisms, prediction, and prevention of ACL injuries. J Orthop Res2016; 34(11): 1843–1855. https://doi.org/10.1002/jor.23414
33.
BelkhelladiMCiersonTMartineauPA. Biomechanical risk factors for increased anterior cruciate ligament loading and injury: a systematic review. Orthop J Sports Med2025; 13(2): 23259671241312681. https://doi.org/10.1177/23259671241312681
34.
RiboliANardiFOstiM, et al. Training load and match locomotor demand in top-class soccer players. J Strength Cond Res2025; 39(2): 249–259. https://doi.org/10.1519/JSC.0000000000004995
KimMCollinsSH. Stabilization of a three-dimensional limit cycle walking model through step-to-step ankle control. In: Proceedings of the IEEE international conference on rehabilitation robot, 2013, p. 6650437. https://doi.org/10.1109/ICORR.2013.6650437
CignettiFSchenaFMottetD, et al. A limit-cycle model of leg movements in cross-country skiing and its adjustments with fatigue. Hum Mov Sci2010; 29(4): 590–604. https://doi.org/10.1016/j.humov.2010.03.001
41.
DelignièresDNourritDDeschampsT, et al. Effects of practice and task constraints on stiffness and friction functions in biological movements. Hum Mov Sci1999; 18(6): 769–793. https://doi.org/10.1016/s0167-9457(99)00040-8
42.
MehdizadehSArshiARDavidsK. A minimal limit-cycle model to profile movement patterns during agility drills. Hum Mov Sci2015; 41: 207–217. https://doi.org/10.1016/j.humov.2015.01.010
43.
MohamedTNicolasBDenisM. A minimal limit-cycle model of hand kinematics in basket-ball-bouncing. Neurosci Lett2005; 382(3): 201–205. https://doi.org/10.1016/j.neulet.2005.03.063
44.
MorinJBBourdinMEdouardP, et al. Mechanical determinants of 100-m sprint running performance. Eur J Appl Physiol2012; 112(11): 3921–3930. https://doi.org/10.1007/s00421-012-2379-8
CainSMMcGinnisRSDavidsonSP, et al. Quantifying performance during load carriage using wireless inertial sensors. Gait Posture2016; 43: 65–69. https://doi.org/10.1016/j.gaitpost.2015.10.016
47.
MadgwickSOHHarrisonAJLVaidyanathanR. Estimation of IMU and MARG orientation using a gradient descent algorithm. In: Proceedings of the IEEE international conference on rehabilitation robot. 2011, pp.1-7. https://doi.org/10.1109/ICORR.2011.5975346
48.
Ruiz-MalagónEJGarcía-PinillosFMolina-MolinaA, et al. RunScribe sacral gait lab validation for measuring pelvis kinematics. Sensors2023; 23(5): 2604. https://doi.org/10.3390/s23052604
49.
SibsonBEBanksJJYawarA, et al. Using inertial measurement units to estimate spine joint kinematics and kinetics during walking and running. Sci Rep2024; 14: 234. https://doi.org/10.1038/s41598-023-50652-w
50.
FaulFErdfelderELangAG, et al. G*Power 3: a flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behav Res Methods2007; 39(2): 175–191. https://doi.org/10.3758/BF03193146
PriscoGPirozziMASantoneA, et al. Validity of wearable inertial sensors for gait analysis: a systematic review. Diagnostics2024; 15(1): 36. https://doi.org/10.3390/diagnostics15010036
KimYKimJMoonJ, et al. Deep learning-based recognition of locomotion mode using inertial measurement units. Sensors2025; 25(4): 1234. https://doi.org/10.3390/s25041234
