Biped Gait Stability Classification Based on the Predicted Step Viability.

In this paper, we address the challenge of ensuring stability in bipedal walking robots and exoskeletons. We explore the feasibility of real-time implementation for the Predicted Step Viability algorithm (PSV), a complex multi-step optimization criterion for planning future steps in bipedal gait. To overcome the high computational cost of the PSV algorithm, we performed an analysis using 11 classification algorithms and a stacking strategy to predict if a step will be stable or not. We generated three datasets of increasing complexity through PSV simulations to evaluate the classification performance. Among the classifiers, k Nearest Neighbors, Support Vector Machine with Radial Basis Function Kernel, Decision Tree, and Random Forest exhibited superior performance. Multi-Layer Perceptron also consistently performed well, while linear-based algorithms showed lower performance. Importantly, the use of stacking did not significantly improve performance. Our results suggest that the feature vector applied with this approach is applicable across various robotic models and datasets, provided that training data is balanced and sufficient points are used. Notably, by leveraging classifiers, we achieved rapid computation of results in less than 1 ms, with minimal computational cost.

Asymmetric effects of different training-testing mismatch types on myoelectric regression via deep learning.

This paper investigates how predictions of a convolutional neural network (CNN) suited for myoelectric simultaneous and proportional control (SPC) are affected when training and testing conditions differ. We used a dataset composed of electromyogram (EMG) signals and joint angular accelerations measured from volunteers drawing a star. This task was repeated multiple times using different combinations of motion amplitude and frequency. CNNs were trained with data from a given combination and tested under different combinations. Predictions were compared between situations in which training and testing conditions matched versus when there was a training-testing mismatch. Changes in predictions were assessed through three metrics: normalized root mean squared error (NRMSE), correlation, and slope of the linear regression between targets and predictions. We found that predictive performance declined differently depending on whether the confounding factors (amplitude and frequency) increased or decreased between training and testing. Correlations dropped as the factors decreased, whereas slopes deteriorated when factors increased. NRMSEs worsened when factors increased or decreased, with more accentuated deterioration for increasing factors. We argue that worse correlations could be related to differences in EMG signal-to-ratio (SNR) between training and testing, which affected the noise robustness of the CNNs' learned internal features. Slope deterioration could be a result of the networks' inability to predict accelerations outside the range seen during training. These two mechanisms may also asymmetrically increase NRMSE. Finally, our findings open further possibilities to develop strategies to mitigate the negative impact of confounding factor variability on myoelectric SPC devices.

A three-dimensional finite element analysis of the sports mouthguard.

BACKGROUND/AIM: The aim of this study was to evaluate the compressive and tensile stresses on dentin and enamel in five different situations: no mouthguard and mouthguards from 1 mm thickness up to 4 mm thickness, using finite element analysis. MATERIALS AND METHODS: A three-dimensional geometry of an upper right central incisor was obtained from a computed tomography and transformed into a mesh separating enamel from dentin. A mouthguard was created covering the buccal surface of the enamel in different thicknesses, and a rubber ball with a velocity of 5 m s(-1) was made as the impact object. RESULTS: The maximum principal stress and the minimal principal stress were evaluated in all situations on dentin and enamel. Both maximum and minimal stress on enamel had the greatest value on the control situation (no mouthguard), and their value decreased as the mouthguard thickness increased. The reduction ranged from 66.62% to 85.5% for compressive stress and from 9.76% to 33.37% for tensile stress on enamel. The results for dentin were similar among the situations with or without mouthguards. CONCLUSION: The mouthguard had beneficial effect considering the stresses on enamel, and between the mouthguard thickness of 3 and 4 mm, there was minimum difference.

Assessment of nose protector for sport activities: finite element analysis.

There has been a significant increase in the number of facial fractures stemming from sport activities in recent years, with the nasal bone one of the most affected structures. Researchers recommend the use of a nose protector, but there is no standardization regarding the material employed. Clinical experience has demonstrated that a combination of a flexible and rigid layer of ethylene vinyl acetate (EVA) offers both comfort and safety to practitioners of sports. The aim of the present study was the investigation into the stresses generated by the impact of a rigid body on the nasal bone on models with and without an EVA protector. For such, finite element analysis was employed. A craniofacial model was constructed from images obtained through computed tomography. The nose protector was modeled with two layers of EVA (1 mm of rigid EVA over 2 mm of flexible EVA), following the geometry of the soft tissue. Finite element analysis was performed using the LS Dyna program. The bone and rigid EVA were represented as elastic linear material, whereas the soft tissues and flexible EVA were represented as hyperelastic material. The impact from a rigid sphere on the frontal region of the face was simulated with a constant velocity of 20 m s(-1) for 9.1 µs. The model without the protector served as the control. The distribution of maximal stress of the facial bones was recorded. The maximal stress on the nasal bone surpassed the breaking limit of 0.13-0.34 MPa on the model without a protector, while remaining below this limit on the model with the protector. Thus, the nose protector made from both flexible and rigid EVA proved effective at protecting the nasal bones under high-impact conditions.

A critical view on biaxial and short-beam uniaxial flexural strength tests applied to resin composites using Weibull, fractographic and finite element analyses.

OBJECTIVE: To evaluate the biaxial and short-beam uniaxial strength tests applied to resin composites based upon their Weibull parameters, fractographic features and stress distribution. METHODS: Disk- (15 mm x 1 mm) and beam-shaped specimens (10 mm x 2 mm x 1 mm) of three commercial composites (Concept/Vigodent, CA; Heliomolar/Ivoclar-Vivadent, HE; Z250/3M ESPE, FZ) were prepared. After 48h dry storage at 37 degrees C, disks and beams were submitted to piston-on-three-balls (BI) and three-point bending (UNI) tests, respectively. Data were analyzed by Weibull statistics. Fractured surfaces were observed under stereomicroscope and scanning electron microscope. Maximum principal stress (sigma(1)) distribution was determined by finite element analysis (FEA). Maximum sigma(1-BI) and sigma(1-UNI) were compared to FZ strengths calculated by applying the average failure loads to the analytical equations (sigma(a-BI) and sigma(a-UNI)). RESULTS: For BI, characteristic strengths were: 169.9a (FZ), 122.4b (CA) and 104.8c (HE), and for UNI were: 160.3a (FZ), 98.2b (CA) and 91.6b (HE). Weibull moduli (m) were similar within the same test. CA and HE presented statistically higher m for BI. Surface pores (BI) and edge flaws (UNI) were the most frequent fracture origins. sigma(1-BI) was 14% lower than sigma(a-BI). sigma(1-UNI) was 43% higher than sigma(a-UNI). SIGNIFICANCE: Compared to the short-beam uniaxial test, the biaxial test detected more differences among composites and displayed less data scattering for two of the tested materials. Also, biaxial strength was closer to the material's strength estimated by FEA.

Stress concentration in microtensile tests using uniform material.

PURPOSE: The objective of this study was to analyze the stress concentration factor (Kt) in specimens of uniform material with the most commonly used geometry (square hourglass) during microtensile tests using finite element analysis. Standardization is emphasized with the aim of obtaining the most representative nominal strength of the material. METHODS: Eighty cases were simulated using three-dimensional models, in which we varied the fixation of specimens in the jig (f = 1 or 2 sides), the height of this fixed region (h = 1 or 2.75 mm), the specimen width (D = 1.5, 2, 3, 4 or 5 mm), and the radius of curvature of the notch (r = 0.2, 0.5, 0.7 or 1 mm). The cross-sectional area (1 mm2) remained constant in all analyses. The stress concentration factor Kt (maximum tensile stress/nominal tensile stress) was calculated. RESULTS: A 150% difference was observed from the lowest Kt value (1.3) to the highest one (3.2). Results indicated that the radius of curvature is a very influential geometric parameter in microtensile strength tests (variation in Kt values up to 47.4%). For two-side fixed specimens, the Kt values varied from 3 to 4%, while the one-side fixed models resulted in variations from 11 to 15%. CONCLUSION: Variations in the specimen geometry and mode of load application can be responsible for part of the different strength values obtained in microtensile tests. The specimen fixation by two sides is a simple and easily performed method to reduce the stress concentration factor and its variations induced by specimen geometry and test assembly.