Fast Human Motion reconstruction from sparse inertial measurement units considering the human shape.

Inertial Measurement Unit-based methods have great potential in capturing motion in large-scale and complex environments with many people. Sparse Inertial Measurement Unit-based methods have more research value due to their simplicity and flexibility. However, improving the computational efficiency and reducing latency in such methods are challenging. In this paper, we propose Fast Inertial Poser, which is a full body motion estimation deep neural network based on 6 inertial measurement units considering body parameters. We design a network architecture based on recurrent neural networks according to the kinematics tree. This method introduces human body shape information by the causality of observations and eliminates the dependence on future frames. During the estimation of joint positions, the upper body and lower body are estimated using separate network modules independently. Then the joint rotation is obtained through a well-designed single-frame kinematics inverse solver. Experiments show that the method can greatly improve the inference speed and reduce the latency while ensuring the reconstruction accuracy compared with previous methods. Fast Inertial Poser runs at 65 fps with 15 ms latency on an embedded computer, demonstrating the efficiency of the model.

Anomaly Detection via Progressive Reconstruction and Hierarchical Feature Fusion.

The main challenges in reconstruction-based anomaly detection include the breakdown of the generalization gap due to improved fitting capabilities and the overfitting problem arising from simulated defects. To overcome this, we propose a new method called PRFF-AD, which utilizes progressive reconstruction and hierarchical feature fusion. It consists of a reconstructive sub-network and a discriminative sub-network. The former achieves anomaly-free reconstruction while maintaining nominal patterns, and the latter locates defects based on pre- and post-reconstruction information. Given defective samples, we find that adopting a progressive reconstruction approach leads to higher-quality reconstructions without compromising the assumption of a generalization gap. Meanwhile, to alleviate the network's overfitting of synthetic defects and address the issue of reconstruction errors, we fuse hierarchical features as guidance for discriminating defects. Moreover, with the help of an attention mechanism, the network achieves higher classification and localization accuracy. In addition, we construct a large dataset for packaging chips, named GTanoIC, with 1750 real non-defective samples and 470 real defective samples, and we provide their pixel-level annotations. Evaluation results demonstrate that our method outperforms other reconstruction-based methods on two challenging datasets: MVTec AD and GTanoIC.

A Hybrid Strategy for Profile Measurement of Micro Gear Teeth.

A hybrid strategy is proposed to meet the challenge of obtaining the profile of micro gear teeth with a small modulus. Firstly, the contact probe segmentally obtained the falling flank profiles with an auxiliary lifting mechanism to avoid interference when it climbs on the rising slope. Then, the noncontact chromatic confocal displacement sensor efficiently acquired the gear peak positions to carry out the two-point error separation with the gear peak positions from the probe measurement. Finally, actual experiments were carried out to obtain the profile of a harmonic drive flexspline. Compared with the commercial ultraprecise profiler, the proposed method provides measurement results with a deviation of less than 20 µm. In conclusion, the hybrid strategy is feasible and accurate for drawing the micro gear teeth profile without any collision between the measuring probes and the measured workpiece.

Enabling Thin-Edged Part Machining of Nomex Honeycomb Composites via Optimizing Variable Angle of Disc Cutters.

Machining Nomex honeycomb composites (NHCs), which are widely-used materials in the aerospace industry, is an imperative process to obtain desired profiles. However, when machining NHCs to obtain a thin-edged surface, some problems can arise due to large cutting forces. To avoid these defects, a method of ultrasonic vibration machining with variable angles of the down milling disc cutter was proposed in this study. The processing principles and motion characteristics of this method were elaborated. A theoretical model of its cutting process was established. The principle of cutting force reduction was qualitatively analyzed based on the model, and an experimental validation was conducted. The results demonstrated that, due to a smaller swing angle in each pass, the proposed method could reduce the fractal dimension of the machined surface by 6.01% compared to 1° with 10° of angle in each pass. And severe machining defects were decreased. Additionally, comparing the process of the fixed 10° angle of ultrasonic vibration machining with the process of a 1° angle in a pass, cutting force can be significantly reduced by 33.5%, demonstrating the effectiveness of the proposed method which improved surface quality by reducing cutting forces.

Fabrication of Optical Fourier Surface by Multiple-Frequency Vibration Cutting for Structural True Coloration.

Optical Fourier surface is a unique patterned optical surface containing the precise sum of sinusoidal waves, each with a well-defined spatial frequency and amplitude. It can manipulate the desired diffracted light field through its Fourier transform, which brings a straightforward mathematical method for designing complex diffractive optics. However, the fabrication techniques typically have the drawbacks of low efficiency, limiting the large-scale industrial application of optical Fourier surfaces. This study presents a powerful approach, the multi-frequency vibration cutting (MFVC), to enable the high-efficiency fabrication of optical Fourier surfaces. A specific optical Fourier surface consisting of arbitrary frequency components of linear gratings has been fabricated on metallic surfaces using MFVC. Due to the capacity of multicomponent gratings in coupling red, green, and blue lights at the same incident angle, the RGB true color has been prepared. The additive and subtractive principles of mixing the three primary colors are demonstrated. The former relies on the light dispersion induced by grating diffraction, while the latter is based on the light absorption induced by the subwavelength grating-coupled surface plasma polarization (SPP). The experimental results of authentic structural true color on the aluminum surface verify the efficacy of MFVC in the fabrication of optical Fourier surfaces.

Effectiveness of Oxygen during Sintering of Silver Thin Films Derived by Nanoparticle Ink.

Silver nanoparticle (NP) inks have been widely used in the ink-jet printing field because of their excellent properties during low-temperature sintering. However, the organic dispersant used to prevent the aggregation and sedimentation of NPs can hinder the sintering process and result in the high resistivity of sintered films. In this study, silver thin films derived from silver NP ink with polyvinylpyrrolidone (PVP) dispersant were sintered in different atmospheres of pure nitrogen, air, and pure oxygen. The effect of the oxygen content in the sintering atmosphere on the thermal properties of the ink, the electrical resistivity and microstructure of the sintered films, and the amount of organic residue were studied by using differential scanning calorimetry, the four-point probe method, scanning electron microscopy, Fourier transform infrared spectroscopy, etc. The mechanism of optimizing the film resistivity by influencing the decomposition of the PVP dispersant and the microstructure evolution of the silver thin films through the sintering atmosphere was discussed. The results demonstrated that an oxygen-containing atmosphere could be effective for silver NPs in two ways. First, the oxygen content could enhance the diffusion ability of silver atoms, thus accelerating the stage transition of microstructural evolution at low temperatures. Second, the oxygen content could enable the PVP to decompose at a temperature much lower than in conditions of pure nitrogen, thus helping to finalize the densification of a silver film with a low resistivity of 2.47 µΩ·cm, which is approximately 1.5-fold that of bulk silver. Our findings could serve as a foundation for the subsequent establishment of ink-jet printing equipment and the optimization of the sintering process for printing silver patterns on flexible substrates.

A temporal-spatial attention-based action recognition method for intelligent fault diagnosis.

The intelligent fault diagnosis of video data has become a demanding task in industrial applications. However, existing models require expensive computational cost and memory demand, which makes this technology applied in factories impossible. To address this problem, a temporal-spatial attention-based action recognition method (TARM) integrating TAB (temporal-attention-based frame splitting model), SAB (spatial-attention-based agent focusing mode) and LSB (long-short term feature learning mode) is proposed. TAB first extracts important frames from raw videos. Then, SAB refines video data by reinforcing their essential features and weakening unnecessary features. Furthermore, LSB monitors action type of video data by establishing recurrent convolutional architectures. Finally, the performance of TARM in terms of training time and fault diagnosis accuracy are validated by comparing with six state-of-the-art video diagnosis methods.

A Fast and Accurate Frequency Tracking Method for Ultrasonic Cutting System via the Synergetic Control of Phase and Current.

Ultrasonic cutting is a superior machining process for brittle materials, owing to its capability to reduce the cutting force and improve the surface quality. To avoid the destructive instability of ultrasonic vibration induced by the cutting force, the excitation frequency of the ultrasonic system must reliably track its resonance frequency. However, it remains challenging for the conventional frequency tracking methods via one parameter to simultaneously achieve both high response rate and high tracking accuracy. This study proposes that more than one parameter could be coupled to get advantages from each parameter. A frequency tracking method via the synergetic control of circuit phase and current of the ultrasonic system was proposed as an example. This method utilizes the phase to responsively determine the tracking direction and uses the characteristic current as the endpoint frequency to ensure accuracy. Theoretical analyses and numerical simulations were conducted to demonstrate that the proposed method can accurately track the frequency of maximum vibration amplitude with a higher response rate than conventional methods. Moreover, ultrasonic cutting tests were performed on Nomex composites to evaluate the machining performance of the ultrasonic system with the proposed method. The experimental results verify that the proposed frequency tracking method enables the ultrasonic system to reach a stable state with a shorter response time, which is beneficial for the reduction of cutting-induced defects.

Investigations on a mathematical model for optimum impedance compensation of a giant magnetostrictive ultrasonic transducer and its resonance characteristics.

Giant magnetostrictive materials (GMMs) have been widely used to fabricate transducers with high-energy output because of their excellent properties. However, there are few reports on mathematical models to optimize the impedance compensation and resonance characteristics of giant magnetostrictive transducers. In this study, a giant magnetostrictive ultrasonic transducer (GMUT) suitable for rotary ultrasonic machining systems is proposed. A mathematical model for optimum impedance compensation that considers the loss in energy conversion is established to maximize the use of ultrasonic energy. The frequency characteristics of the electrical feedback signal in the resonance state are investigated, and the resonance zone found is used for frequency tracking. An impedance analyzer is used to determine the parameters of the mathematical model, and the validity of the optimum compensation capacitance is verified by experiments. The frequency characteristics of the minimum current, active power, and amplitude are obtained to obtain the resonance zone in the GMUT with the lowest energy consumption. The results of this study provide a reference for frequency tracking.

An amplitude prediction model for a giant magnetostrictive ultrasonic transducer.

Rotary ultrasonic machining (RUM) is widely used in the processing of brittle and hard materials. The application of giant magnetostrictive ultrasonic transducers (GMUT) with effective vibration performance is an increasingly popular field of research within RUM. A generalized amplitude prediction model for GMUT is obtained in this paper by first providing an equivalent kinetics model of the GMUT. Considering the influence on interaction force between Terfenol-D and the external mechanical mechanism, the prestress mechanism of Terfenol-D and the joint face of the horn are determined as equivalent to two spring-damping systems in series, and a general GMUT vibration equation is established. The equivalent stiffness of the prestress mechanism is then identified, and the mechanical quality factor of the vibration system is calculated by impedance analysis. The influence of the joint face of the horn and the prestress mechanism on the amplitude is then studied by nonlinear least square fitting. Based on a magnetostriction and magnetization model, an odd power amplitude prediction model with mechanical quality factor, excitation current amplitude, and excitation frequency is proposed. The experimental results demonstrate that the proposed model can effectively predict the output amplitude of the GMUT with different mechanical quality factors for different excitation signals, providing a method for system design and optimization of the GMUT.

Building of Longitudinal Ultrasonic Assisted Turning System and Its Cutting Simulation Study on Bulk Metallic Glass.

Bulk metallic glass (BMG) is a new kind of material which is made by rapid condensation of alloy. With excellent properties like high strength, high hardness, corrosion resistance, BMG is increasingly applied in mold manufacturing, weapon equipment and other fields. However, BMG is also one of hard-to-machine materials, which is arduous to be processed precisely and efficiently by the means of conventional cutting. Compared with conventional cutting, ultrasonic machining has a multitude of technological advantages such as reducing the cutting force, extending the tool life, etc. In ultrasonic machining, the ultrasonic electric signal is transformed into high frequency mechanical vibration on the tool, which changes the relationship between the tool and the workpiece in the process of machining. In this study, the longitudinal ultrasonic assisted turning (LUAT) system is established for processing BMG. Its resonant frequency and vibration characteristics are first simulated by modal analysis and harmonic response analysis, and then tested by displacement testing experiments, so that the suitable frequency and the amplitude for BMG turning can be selected and verified. On this basis, the two-dimensional turning finite element model is established to study the effect of ultrasonic vibration on cutting force under different cutting speeds. The research manifest that during the BMG turning, the assistance of longitudinal ultrasonic vibration can significantly reduce the average cutting force as well as the von Mises stress when the turning speed is below the critical turning speed. In addition, the tip of the tool contacts the workpiece discontinuously during cutting process which makes the instantaneous turning force in LUAT more periodic than that in conventional turning (CT).

Investigations on the critical feed rate guaranteeing the effectiveness of rotary ultrasonic machining.

Rotary ultrasonic machining (RUM) is a well-known and efficient method for manufacturing holes in brittle materials. RUM is characterized by improved material removal rates, reduced cutting forces and reduced edge chipping sizes at the hole exit. The aim of this study is to investigate the critical feed rate to guarantee the effectiveness of RUM. Experimental results on quartz glass and sapphire specimens show that when the feed rate exceeds a critical value, the cutting force increases abruptly, accompanied by a significant decrease of ultrasonic amplitude. An analytical model for the prediction of critical feed rates is presented, based on indentation fracture mechanic and the theory of impact of vibrating systems. This model establishes the theoretical relationships between the critical feed rate, idling resonant ultrasonic amplitude and spindle speed. The results predicted by the analytical model were in good agreement with the experimental results.

Comparison of wear behaviors for an artificial cervical disc under flexion/extension and axial rotation motions.

The wear behaviors of a ball-on-socket (UHMWPE-on-Ti6Al4V) artificial cervical disc were studied with 1.5 MC (million cycles) wear simulation under single flexion/extension and axial rotation motion and their composite motion. The wear rates, wear traces, and contact stress were analyzed and contrasted based on mass loss, optical microscopy and SEM as well as 3D profilometer, and ANSYS software, respectively. A much higher wear rate and more severe wear scars appeared under multi-directional motion. Flexion/extension motion of 7.5° lead to more severe wear than that under axial rotation motion of 4°. The above results were closely related to the contact compression stress and shear stress. The wear surface in FE motion showed typical linear wear scratches while revealing obvious arc-shaped wear tracks in AR motion. However, the central zone of both ball and socket components revealed more severe wear tracks than that in the edge zone under these two different motions. The dominant wear mechanism was plowing/scratching and abrasive wear as well as a little oxidation wear for the titanium socket while it was scratching damage with adhesive wear and fatigue wear due to plastic deformation under cyclic load and motion profiles for the UHMWPE ball.