Motion Curves Research Articles

SKEL-Betweener: a Neural Motion Rig for Interactive Motion Authoring

Authoring 3D motions is a laborious process that requires manipulating and coordinating many control handles over time. Neural motion representations learned from large motion datasets have recently shown impressive capabilities in many motion completion tasks. However, current methods are not designed for interactive motion authoring workflows. The reasons being their requirement of a dense context of full poses, which takes considerable time to author, as well as their lack of joint-level controls for refinement. In this paper, we introduce a Neural Motion Rig called SKEL-Betweener, tailored to interactive motion authoring. SKEL-Betweener is able to generate long motion sequences from two poses only, and enables intermediate motion authoring via neural motion curves---intuitive joint-level controls for positions and orientations. Through user evaluations, we demonstrate the effectiveness of our Neural Motion Rig for efficiently creating and editing motions.

Research on magnetic nonchain transport of steel cartridge casing ammunition

Magnetic force is used to drive ferromagnetic objects by shortening the magnetic flux path to reduce magnetic reluctance and increase magnetic conductivity. Based on this characteristic, magnetic force drives have been widely applied in various fields, such as military, transportation, and engineering, attracting significant attention. This paper proposes a method for the nonchain transport of steel shell ammunition based on magnetic force. The force generated by the minimal magnetic resistance in the air gap is utilized to drive the steel shell ammunition. A calculation model for the driving component is proposed using the Maxwell stress tensor method, and the accuracy of the model is verified by simulating ammunition motion curves using finite element software. Finally, prototype experiments are conducted to explore the motion conditions under uncertain ammunition masses. The magnetic nonchain ammunition transport mechanism features a simpler structure and lower maintenance costs, making it more suitable for use on unmanned platforms.

Effect of asymptomatic intervertebral flexion patterns on lumbar disc pressure: A finite element analysis study.

Movement patterns may be a factor for manipulating the lumbar load, although little information is yet available in the literature about the relationship between this variable and intervertebral disc pressure (IDP). A finite element model of the lumbar spine (49-year-old asymptomatic female) was used to simulate intervertebral movements (L2-L5) of 127 asymptomatic participants. The data from participants that at least completed a simulation of lumbar vertebral movement during the first 53% of a movement cycle (flexion phase) were used for further analyses. Then, for each vertebral angular motion curve with constant spatial peaks, different temporal patterns were simulated in two stages: (1) in lumbar pattern exchange (LPE), each vertebral angle was simulated by the corresponding vertebrae of other participants data; (2) in vertebral pattern exchange (VPE), vertebral angles were simulated by each other. The k-mean algorithm was used to cluster two groups of variables; peak and cumulative IDP, in both stages of simulations (i.e., LPE and VPE). In the second stage of the simulation (VPE), Kendall's tau was utilized to consider the relationship between different temporal patterns and IDPs for each individual lumbar level. Cluster analyses showed that the temporal movement pattern did not exhibit any effect on the peak IDP while the cumulative IDP changed significantly for some patterns. Earlier involvement in lumbar motion at any level led to higher IDP in the majority of simulations. There is therefore a possibility of manipulating lumbar IDP by changing the temporal pattern with the same ROM, in which optimal distribution of the loads among lumbar levels may be applied as preventive or treatment interventions. Evaluating load benefits, such as load, on biomechanically relevant lumbar levels, dynamically measured by quantitative fluoroscopy, may help inform interventional exercises.

Apsidal motion and TESS light curves of three southern close eccentric eclipsing binaries: GM Nor, V397 Pup, and PT Vel

The study of apsidal motion in eccentric eclipsing binaries provides an important observational test of theoretical models of stellar structure and evolution. New ground-based and space-based photometric data have been obtained and archival spectroscopic measurements were used in this study of three detached early-type and southern-hemisphere eccentric eclipsing binaries GM Nor (P = 1d​​.88, e = 0.05), V397 Pup (3d​​.00, 0.30), and PT Vel (1d​​.80, 0.12). Their TESS observations in several sectors have also been included and the corresponding light curves were solved using the PHOEBE code. As a result, new accurate photoelectric times of minimum light have been obtained. The newly completed O − C diagrams were analyzed using all reliable timings found in the literature and calculated using the TESS light curves. New or improved values for the elements of apsidal motion were obtained. Using ESO archive spectroscopy, for V397 Pup, the precise absolute parameters were newly derived: M1 = 3.076(35) M⊙, M2 = 2.306(35) M⊙, and R1 = 2.711(55) R⊙, R2 = 1.680(55) R⊙. For PT Vel the absolute dimensions were improved: M1 = 2.204(25) M⊙, M2 = 1.638(25) M⊙, and R1 = 2.108(30) R⊙, R2 = 1.605(30) R⊙. For GM Nor, the less accurate absolute parameters based on the light curve analysis were evaluated: M1 = 1.94(15) M⊙, M2 = 1.84(14) M⊙, and R1 = 2.27(20) R⊙, R2 = 2.25(20) R⊙. We found more precise and relatively short periods of apsidal motion of about 80, 335, and 160 years, along with the corresponding internal structure constants, log k2, –2.524, –2.361, and –2.563, for GM Nor, V397 Pup, and PT Vel, respectively. Relativistic effects are small but not negligible, making up to 10% of the total apsidal motion rate in all systems. No marks of the presence of the third body were revealed in the light curves, on the O − C diagrams, or in the reduced spectra of the eccentric systems studied here.

Research on the Motion Control Strategy of a Lower-Limb Exoskeleton Rehabilitation Robot Using the Twin Delayed Deep Deterministic Policy Gradient Algorithm.

The motion control system of a lower-limb exoskeleton rehabilitation robot (LLERR) is designed to assist patients in lower-limb rehabilitation exercises. This research designed a motion controller for an LLERR-based on the Twin Delayed Deep Deterministic policy gradient (TD3) algorithm to control the lower-limb exoskeleton for gait training in a staircase environment. Commencing with the establishment of a mathematical model of the LLERR, the dynamics during its movement are systematically described. The TD3 algorithm is employed to plan the motion trajectory of the LLERR's right-foot sole, and the target motion curve of the hip (knee) joint is deduced inversely to ensure adherence to human physiological principles during motion execution. The control strategy of the TD3 algorithm ensures that the movement of each joint of the LLERR is consistent with the target motion trajectory. The experimental results indicate that the trajectory tracking errors of the hip (knee) joints are all within 5°, confirming that the LLERR successfully assists patient in completing lower-limb rehabilitation training in a staircase environment. The primary contribution of this study is to propose a non-linear control strategy tailored for the staircase environment, enabling the planning and control of the lower-limb joint motions facilitated by the LLERR.

Propagation Laws of Ultrasonic Continuous Signals at the Transmitting Transducer–Soil Interface

Ultrasonic detection is one of the main methods for information detection and has advantages in soil detection. Ultrasonic signals attenuate in soil, resulting in unique propagation laws. This paper studies the propagation laws of ultrasound in soil, focusing on the propagation characteristics of ultrasonic continuous signals at the transducer–soil interface. This study uses excitation frequency and amplitude as experimental factors and employs the discrete element simulation method to analyze the vibration characteristics of soil particles. It reveals the relationship between changes in soil pressure at the interface and the movement of the transducer. The results show that the motion curve of the transmitting transducer lags behind the soil pressure changes, and the energy of the ultrasonic signal increases with higher excitation frequency and amplitude. Specifically, the peak value of the first wave |H0| at 40 kHz and 60 kHz is 210% and 263% of that at 20 kHz, respectively. When the excitation amplitude increases from 0.005 mm to 0.015 mm, the value of the peak value of other waves |H| increases by 323%. This paper preliminarily reveals the propagation laws of ultrasonic continuous signals at the transducer–soil interface, providing theoretical support for the development of ultrasonic soil property detection instruments.

Large-stroke and high-precision coaxial integrated macro-micro composite actuator based on VCM and GMA

A coaxially integrated macro-micro composite actuator (MMCA) with large stroke and high accuracy is proposed by combining a voice coil motor (VCM) with the giant magnetostrictive actuator (GMA). The magnetic circuit model of the macro-motion part is established based on the driving principle of VCM, and the multi-field coupling model of the micro-motion part is established based on the Jiles-Atherton model. The finite element method was used to analyze the relationship between displacement, output force, velocity, acceleration, and time of the macro-motion part under different currents, the magnetic flux density, output force, and displacement curves of the micro-motion part, and the mutual influence between the macro and micro motion parts were analyzed. The prototype of the MMCA was developed, and an experimental test platform was built. The results show that the MMCA macro-motion displacement curve can fit the simulation curve well during open-loop positioning, and when the current size of the input macro-motion coil is 4 A, the experimental curve of the MMCA is the most consistent with the simulation curve. When closed-loop control, the motion curve of the drive can well follow the set displacement curve, in which the maximum stroke of the prototype developed is 50 mm, the positioning error of the macro-motion part is less than 20 μm, the maximum stroke of the micro-motion part is 40 μm, and the overall positioning accuracy of the MMCA is 0.14 μm. The research results provide a new idea and theoretical basis for further optimization and development of precision positioning platform with high precision and large stroke.

Mathematical model of fatigue degradation of cylindrical piezo element of linear micro piezo motor

In the article, a mathematical model has been constructed and investigated by analyzing the problem of disintegration during torsion of the cylindrical shaped piezo element in the linear piezoelectric engine. Nonlinear differential equations expressing the expansion of the disintegration surface formed during the torsion of the piezo element under the influence of variable torsional moment have been formulated. Guidelines determining the initial degradation period for the development of the disintegration process have been obtained, and based on the developed mathematical model, displacement curves of the disintegration surface motion have been established.

Design, modeling, and experiment of a multi-degree-of-freedom pneumatic soft bionic actuator

Drawing on the driving mechanism of biological muscles and combining the nonlinear hyperelastic characteristics of silicone rubber, a multi-degree-of-freedom pneumatic soft bionic actuator is designed, which can be used as the executing mechanism for soft robots and robotic arms. Using response surface analysis and numerical simulation algorithms, the optimal combination of structural dimensions parameters is determined with the maximum bending and torsion angles output by the actuator as the optimization objectives. Based on the idea of flexible mechanism Piecewise Constant Curvature (PCC) modeling, the kinematics equation of the actuator is derived. Analyze the equivalent motion structure of the actuator and establish a dynamic model of the actuator based on the Lagrange method. The physical model of the actuator is manufactured using rapid prototyping technology. Finally, experimental testing and analysis are conducted on the physical model to obtain the motion and dynamic characteristics curves and empirical formulas of the actuator, which are compared with theoretical and simulation results to verify that the pneumatic soft bionic actuator is feasible and effective. The above research methods, processes, and results can provide reference and inspiration for the research and implementation of pneumatic and hydraulic soft robots and gripping actuators for soft robotic arms.

Robot Motion Planning Based on an Adaptive Slime Mold Algorithm and Motion Constraints

The rapid advancement of artificial intelligence technology has significantly enhanced the intelligence of mobile robots, facilitating their widespread utilization in unmanned driving, smart home systems, and various other domains. As the scope, scale, and complexity of robot deployment continue to expand, there arises a heightened demand for enhanced computational power and real-time performance, with path planning emerging as a prominent research focus. In this study, we present an adaptive Lévy flight–rotation slime mold algorithm (LRSMA) for global robot motion planning, which incorporates LRSMA with the cubic Hermite interpolation. Unlike traditional methods, the algorithm eliminates the need for a priori knowledge of appropriate interpolation points. Instead, it autonomously detects the convergence status of LRSMA, dynamically increasing interpolation points to enhance the curvature of the motion curve when it surpasses the predefined threshold. Subsequently, it compares path lengths resulting from two different objective functions to determine the optimal number of interpolation points and the best path. Compared to LRSMA, this algorithm reduced the minimum path length and average processing time by (2.52%, 3.56%) and (38.89%, 62.46%), respectively, along with minimum processing times. Our findings demonstrate that this method effectively generates collision-free, smooth, and curvature-constrained motion curves with the least processing time.

Analysis of Horizontal Cylinder Load under Different Conditions in Regards to Waves and Flows

A numerical simulation based on the CFD method is used to study the interaction between a horizontal cylinder and wave flow. Firstly, a two-dimensional numerical calculation model of both a fixed and a rigid moving cylinder, with a free surface under varying wave flow conditions, is created. In the established model, the loads on the horizontal cylinder under different submergence depths, flow velocities, cylinder sizes, wave periods, and k values (spring stiffness) are analyzed and calculated. The results show that, when the cylinder is close to the free surface, its hydrodynamic load under wave flow conditions is more sensitive to changes in submergence depth, which essentially affects wave reflection and blockage. At different flow velocities, k values, cylinder radii, and arm lengths, the main frequency of the Fourier transform of the cylinder motion curve remains unchanged; however, the main frequency does change with the wave period and submergence depth. The efficiency of rotary cylindrical energy harvesting is influenced by various factors, among which an initial increase and then decrease are observed with a gradually increasing k value, arm length, period, and radius, in addition to an observed decrease with increasing flow velocity.

Monitoring Cattle Ruminating Behavior Based on an Improved Keypoint Detection Model.

Cattle rumination behavior is strongly correlated with its health. Current methods often rely on manual observation or wearable devices to monitor ruminating behavior. However, the manual monitoring of cattle rumination is labor-intensive, and wearable devices often harm animals. Therefore, this study proposes a non-contact method for monitoring cattle rumination behavior, utilizing an improved YOLOv8-pose keypoint detection algorithm combined with multi-condition threshold peak detection to automatically identify chewing counts. First, we tracked and recorded the cattle's rumination behavior to build a dataset. Next, we used the improved model to capture keypoint information on the cattle. By constructing the rumination motion curve from the keypoint information and applying multi-condition threshold peak detection, we counted the chewing instances. Finally, we designed a comprehensive cattle rumination detection framework to track various rumination indicators, including chewing counts, rumination duration, and chewing frequency. In keypoint detection, our modified YOLOv8-pose achieved a 96% mAP, an improvement of 2.8%, with precision and recall increasing by 4.5% and 4.2%, enabling the more accurate capture of keypoint information. For rumination analysis, we tested ten video clips and compared the results with actual data. The experimental results showed an average chewing count error of 5.6% and a standard error of 2.23%, verifying the feasibility and effectiveness of using keypoint detection technology to analyze cattle rumination behavior. These physiological indicators of rumination behavior allow for the quicker detection of abnormalities in cattle's rumination activities, helping managers make informed decisions. Ultimately, the proposed method not only accurately monitors cattle rumination behavior but also provides technical support for precision management in animal husbandry, promoting the development of modern livestock farming.

Diagnostic efficacy of complexity metrics from cardiac MRI myocardial segmental motion curves in detecting late gadolinium enhancement in myocardial infarction patients

BackgroundMyocardial segmental motion is associated with cardiovascular pathology, often assessed through myocardial strain features. The stability of the motion can be influenced by myocardial fibrosis. This research aimed to explore the complexity metrics (CM) of myocardial segmental motion curves, observe their correlation with late gadolinium enhancement (LGE) transmural extension (TE), and assess diagnostic efficacy combined with segmental strains in different TE segments. MethodsWe included 42 myocardial infarction patients, dividing images into 672 myocardial segments (208 remote, 384 viable, and 80 unviable segments based on TE). Radial and circumferential segmental strain, along with CM for motion curves, were extracted. Correlation between CM and LGE, as well as the potential distinguishing role of CM, was evaluated using Pearson correlation, univariate linear regression (F-test), multivariate regression analysis (T-test), area under curve (AUC), machine learning models, and DeLong test. ResultsAll CMs showed significant linear correlation with TE (P < 0.001). Six CMs were correlated with TE (r > 0.3), with radial frequency drift (FD) displayed the strongest correlation (r = 0.496, P < 0.001). Radial and circumferential FD significantly differed in higher TE myocardium than in remote segments (P < 0.05). Radial FD had practical diagnostic efficacy (remote vs. unviable AUC = 0.89, viable vs. unviable AUC = 0.77, remote vs. viable AUC = 0.65). Combining CM with segmental strain features boosted diagnostic efficacy than models using only segmental strain features (DeLong test, P < 0.05). ConclusionsThe CM of myocardial motion curves has been associated with LGE infarction, and combining CM with strain features improves the diagnosis of different myocardial LGE infarction degrees.

Key Technology Research and Equipment Development of Automatic Spool Feeding Device

To solve the problems of high labour intensity and low efficiency of artificial spool feeding, an automatic spool feeding device is developed. SolidWorks is used for overall structural scheme design and determination of design parameters. The kinematic equation of the chain conveyor line is established, a mathematical model is solved with MATLAB, and sprocket, and chain parameters are selected. The clamping mechanism and turnover mechanism are designed, theoretical analysis of the key structure is completed. Dynamic simulation of the chain conveyor line is carried out by ADAMS, and preliminary operational stability is determined by combining motion curves. The test prototype is developed for feeding tests and measurement of the operating parameters. The test results show that the efficiency of machine feeding is 13% higher than that of manual feeding, the average feeding time is 43% lower, the halt waiting time is 43% lower, and the operating parameters are by the theoretical design and the work is reliable. The research results can provide a reference for similar spool-feeding devices.

Linearizing the vertical scale of an interferometric microscope and its effect on step-height measurement.

The vertical scale calibration of an interferometric microscope is important for establishing traceability of surface topography measurements to the International System of Units (SI) unit of length, the meter. Building on the calibration procedure for the amplification coefficient developed by de Groot and Beverage [Proc. SPIE 9526, 952610 (2015)], this paper describes a calibration procedure that yields the response curve for the entire vertical scan motion of a coherent scanning interferometric microscope. The method requires only a flat mirror as an artifact, a narrow band spectral filter, an aperture to reduce the effective numerical aperture, and the ability to raise and lower the microscope head so that the center of the interferogram can be varied within the scan range. The local frequency of the interferogram is determined by fitting sections of the interferogram to a sinusoidal function. The nonlinearity determined from the local frequency data can be used to estimate the uncertainty in uncorrected vertical height measurements. We describe how optical profile data can be corrected for nonlinearity due to dynamic effects in the scan motion and show that the correction improves the reproducibility of step height measurements by at least a factor of three and close to that of the repeatability.

Clinical and Radiological Predictors of Slower and Non-Improvement Following Surgical Treatment of L4-5 Degenerative Spondylolisthesis: Preliminary Results.

Retrospective cohort. To identify the predictors of slower and non-improvement following surgical treatment of L4-5 degenerative lumbar spondylolisthesis (DLS). There is limited evidence regarding clinical and radiological predictors of slower and non-improvement following surgery for L4-5 DLS. Patients who underwent minimally invasive decompression or fusion for L4-5 DLS and had a minimum of 1-year follow-up were included. Outcome measures were: (1) minimal clinically important difference (MCID), (2) patient acceptable symptom state (PASS), and (3) global rating change (GRC). Clinical variables analyzed for predictors were age, gender, body mass index (BMI), surgery type, comorbidities, anxiety, depression, smoking, osteoporosis, and preoperative patient reported outcome measures (PROMs) (Oswestry Disability Index, ODI; Visual Analog Scale, VAS back and leg; 12-Item Short Form Survey Physical Component Score, SF-12 PCS). Radiological variables analyzed were slip percentage, translational and angular motion, facet diastasis/cyst/orientation, laterolisthesis, disc height, scoliosis, main and fractional curve Cobb angles, and spinopelvic parameters. 233 patients (37% decompression, 63% fusion) were included. At <3 months, high pelvic tilt (PT) (OR 0.92, P 0.02) and depression (OR 0.28, P 0.02) were predictors of MCID non-achievement and GRC non-betterment, respectively. Neither retained significance at >6 months and hence, were identified as predictors of slower improvement. At >6 months, low preoperative VAS leg (OR 1.26, P 0.01) and high facet orientation (OR 0.95, P 0.03) were predictors of MCID non-achievement, high L4-5 slip percentage (OR 0.86, P 0.03) and L5-S1 angular motion (OR 0.78, P 0.01) were predictors of GRC non-betterment, and high preoperative ODI (OR 0.96, P 0.04) was a predictor of PASS non-achievement. High PT and depression were predictors of slower improvement and low preoperative leg pain, high disability, high facet orientation, high slip percentage, and L5-S1 angular motion were predictors of non-improvement. However, these are preliminary findings and further studies with homogeneous cohorts are required to establish these findings.

An Integrated C4-Spline Interpolation and Time-Free Global Optimization Methodology Applied to High-Speed Cam Motion Design

The optimal tuning of high-order motion parameters has emerged as a promising strategy for actively controlling the kinematics/dynamics of high-speed cam mechanisms. However, accomplishing this task remains challenging with current low-order interpolation or tuning methods. This study proposes an integrated high-order interpolation and tuning methodology for the optimal construction of high-speed motion curves. Initially, an explicit C4-spline interpolant (C4SI) is developed. This interpolant utilizes four-order continuous (C4) splines to synthesize a high-fidelity motion curve that satisfies the predefined motion constraints up to the fourth order, including dimensionless displacement, velocity, acceleration, jerk, and quirk. Concerning the reduction of motion peaks, a unique C4SI-based global kinematics optimization strategy is designed, using the definite integral of the motion curve (free of the time variable) as the objective function. This facile time-free optimization strategy could yield a simultaneous reduction in multiple motion peaks (up to five), which is currently inaccessible with conventional motion tuning strategies. Concerning the improvement of dynamic characteristics, the C4SI-based time-free global dynamics optimization of variable motion parameters is further performed. The results indicate that the optimized fourth-order motion curve offers minimal high-speed transmission error and residual vibration over the whole rise-dwell-return-dwell cycle, which outperforms the standard motion curves and other low-order counterparts.

Design and Simulation of Exoskeleton Robot

Based on the research of lower limb exoskeleton robots and extensive literature review, combined with human anatomy and kinematics, a four-degree-of-freedom lower limb exoskeleton robot was designed, and its three-dimensional model was established using UG software. In the design of the exoskeleton robot, it is required to have a simple structure, zero sensitivity in response speed, and be easy to realize automatic control. The exoskeleton robot designed in this paper uses electric push rods as the driving elements, and the dimensions of each part of the robot are obtained from the national standard human body size table; based on the analysis of the human gait cycle, and using cubic polynomial interpolation to plan the spatial posture of each joint; Adams was used to simulate the movement of the exoskeleton robot, and the motion curves of each joint were obtained.

Coupled surface diffusion and mean curvature motion: An axisymmetric system with two grains and a hole

Thin polycrystalline solid state films, which are used in many technological applications, can exhibit various phenomena, such as wetting, dewetting, and hole formation. We focus on a model system containing two contacting grains which surround a hole. For simplicity, the system is assumed to be axisymmetric, to be supported by a planar substrate and to be bounded within an inert semi-infinite cylinder. We assume that the exterior surfaces of the grains evolve by surface diffusion and the grain boundary between the adjacent grains evolve by motion by mean curvature. Boundary conditions are imposed following W.W. Mullins, 1958. Parametric formulas are derived for the steady states, which contain two nodoids describing the exterior surfaces, which are coupled to a catenoid which describes the grain boundary. At steady state, the physical parameters of the system may be prescribed via two angles,β\beta, the angle between the exterior surface and the grain boundary, andθc\theta _c, the contact angle between the exterior surface and the substrate; additionally, there are two dimensionless geometric parameters which must satisfy certain constraints. We prove that ifβ∈(π/2,π)\beta \in (\pi /2, \pi )andθc=π\theta _c=\pi, then there exists a continuum of steady states. Numerical calculations indicate that steady state profilescan exhibit physical features, such as hillock formation; a fuller numerical study of the steady states and their properties recently appeared in Zigelman and Novick-Cohen [J. Appl. Phys. 134 (2023), 135302], which relies on the formulas and results derived here.

Feasibility of predicting functional decline in the elderly through key posture information during sit-to-stand movement

BackgroundEarly detection of functional decline in the elderly in day care centres facilitates timely implementation of preventive and treatment measures. Research questionWhether or not a predictive model can be developed by applying image recognition to analyze elderly individuals' posture during the sit-to-stand (STS) manoeuvre. MethodsWe enrolled sixty-six participants (24 males and 42 females) in an observational study design. To estimate posture key point information, we employed a region-based convolutional neural network model and utilized nine key points and their coordinates to calculate seven eigenvalues (X1-X7) that represented the motion curve features during the STS manoeuvre. One-way analysis of variance was performed to evaluate four STS strategies and four types of compensation strategies for three groups with different capacities (college students, community-dwelling elderly, and day care center elderly). Finally, a machine learning predictive model was established. ResultsSignificant differences (p < 0.05) were observed in all eigenvalues except X2 (momentum transfer phase, p = 0.168) between participant groups; significant differences (p < 0.05) were observed in all eigenvalues except X2 (p = 0.219) and X3 (hip-rising phase, p = 0.286) between STS patterns; significant differences (p < 0.05) were observed in all eigenvalues except X2 (p = 0.842) and X3 (p = 0.074) between compensation strategies. The motion curve eigenvalues of the seven posture key points were used to build a machine learning model with 85% accuracy in capacity detection, 70% accuracy in pattern detection, and 85% accuracy in compensation strategy detection. SignificanceThis study preliminarily demonstrates that eigenvalues can be used to detect STS patterns and compensation strategies adopted by individuals with different capacities. Our machine learning model has excellent predictive accuracy and may be used to develop inexpensive and effective systems to help caregivers to continuously monitor STS patterns and compensation strategies of elderly individuals as warning signs of functional decline.