Velocity Trajectory Research Articles

On Uniform Null Controllability of Transport–Diffusion Equations With Vanishing Viscosity Limit

ABSTRACTThis paper aims to address an interesting open problem, posed in the paper “Singular Optimal Control for a Transport‐Diffusion Equation” of Sergio Guerrero and Gilles Lebeau in 2007. The problem involves studying the null controllability cost of a transport–diffusion equation with Neumann conditions, where the diffusivity coefficient is denoted by and the velocity by . Our objective is twofold. First, we investigate the scenario where each velocity trajectory originating from enters the control region in a shorter time at a fixed entry time. By employing Agmon and dissipation inequalities, and Carleman estimate in the case is the gradient of a time‐dependent scalar field, we establish that the control cost remains bounded for sufficiently small and large control time. Secondly, we explore the case where at least one trajectory fails to enter the control region and remains in . In this scenario, we prove that the control cost explodes exponentially when the diffusivity approaches zero and the control time is sufficiently small for general velocity.

Deep Temporal Clustering of Pathological Gait Patterns in Post-Stroke Patients Using Joint Angle Trajectories: A Cross-Sectional Study.

Rehabilitation of gait function in post-stroke hemiplegic patients is critical for improving mobility and quality of life, requiring a comprehensive understanding of individual gait patterns. Previous studies on gait analysis using unsupervised clustering often involve manual feature extraction, which introduces limitations such as low accuracy, low consistency, and potential bias due to human intervention. This cross-sectional study aimed to identify and cluster gait patterns using an end-to-end deep learning approach that autonomously extracts features from joint angle trajectories for a gait cycle, minimizing human intervention. A total of 74 sub-acute post-stroke hemiplegic patients with lower limb impairments were included in the analysis. The dataset comprised 219 sagittal plane joint angle and angular velocity trajectories from the hip, knee, and ankle joints during gait cycles. Deep temporal clustering was employed to cluster them in an end-to-end manner by simultaneously optimizing feature extraction and clustering, with hyperparameter tuning tailored for kinematic gait cycle data. Through this method, six optimal clusters were selected with a silhouette score of 0.2831, which is a relatively higher value compared to other clustering algorithms. To clarify the characteristics of the selected groups, in-depth statistics of spatiotemporal, kinematic, and clinical features are presented in the results. The results demonstrate the effectiveness of end-to-end deep learning-based clustering, yielding significant performance improvements without the need for manual feature extraction. While this study primarily utilizes sagittal plane data, future analysis incorporating coronal and transverse planes as well as muscle activity and gait symmetry could provide a more comprehensive understanding of gait patterns.

Theoretical Aspects of Soil Layer Turnover Within the Boundaries of Its Own Furrow

The paper highlights that soil layer turnover remains the most widely used method of primary tillage. Among the existing techniques, smooth plowing without producing back ridges or furrows, which is achieved using reversible plows, best meets the high standards of modern agricultural practices. (Research purpose) The study aims to substantiate the kinematics of soil layer turnover within the boundaries of its own furrow without lateral displacement. (Materials and methods) In analyzing the kinematics, the soil layer is assumed to behave as a cohesive elastic substance undergoing deformation during turnover within its own furrow, without disintegration. This assumption is quite reasonable, as it is well-established that a sodded and moist layer can be extracted as a continuous, intact strip that retains its geometric dimensions when turned 180°. The trajectory of the layer is derived using classical methods of theoretical mechanics. (Results and discussions) The equations governing the motion of the soil layer points during turnover within its own furrow are analyzed. During this process, all points in the cross-section of the layer undergo spatial displacement. Changes in displacement, velocity, and acceleration of the i-th point of the hypothetical layer exhibit smooth dependencies described by trigonometric functions. However, at a rotation angle of ωt = π/2, an abrupt change occurs in the trajectories of displacement, velocity and acceleration graphs, indicating sharply variable loads acting on the soil layer at this point. The abrupt change is attributed to a shift in the support rib which serves as the axis for the rotation of the soil layer's crosssection. The center of gravity of the cross-section moves with variable velocity and acceleration, which indicates the presence of inertial forces. Overcoming these forces requires a certain amount of energy. (Conclusions) The energy required largely depends on the geometric parameters of the layer a, b and its rotation modes (ω). When the layer cross-section rotates by an angle When the layer cross-section rotates by an angle ωt = π/2 – γ, the vertical acceleration of the central point (О) reaches its maximum value. Under certain conditions, the soil layer may detach from the furrow bottom in this position. The kinematics analysis of a soil layer turnover within the boundaries of its own furrow reveals new phenomena occurring during its motion to identifies the patterns of influence that the soil layer's geometric parameters exert on its dynamic characteristics.

Modeling Supply Air Jet of Split Air Conditioner Based on Abramovich Jet Theory and Bayesian Markov Chain Monte Carlo Algorithm

Aim: This study aims to develop supply air jet models for split air conditioners. Background: Designing and operating an air conditioning system based on the unique characteristics of the supply air jet is essential to control the draft risk. Motivation: Due to the distinct differences in air supply characteristics of the split air conditioners caused by the intermittent operation and the complex design of supply air outlets, the existing supply air jet models designed for central air conditioning systems are unsuitable for split air conditioners. Significant results: In this study, models for the trajectory position, velocity, and temperature of the supply air jet for split air conditioners are established based on the Abramovich jet theory and Bayesian Markov Chain Monte Carlo (MCMC) algorithm. The effectiveness of the proposed models is demonstrated via comparison with the training dataset from the experimentally validated Computational Fluid Dynamics simulation of a split air conditioner. For the model testing dataset, the RMSEs of trajectory position, velocity, and temperature are 0.01 m, 0.09 m/s, and 0.09 °C, respectively, with a low supply air velocity and 0.01 m, 0.25 m/s, and 0.03 °C with a high supply air velocity. The proposed models of velocity and temperature of the supply air jet are general for various supply air conditions, while coefficients in the trajectory position model of the supply air jet need to be recalibrated with the proposed Bayesian MCMC algorithm for different supply air conditions. The proposed models provide valuable guidance for the operation of split air conditioners.

Real-time position path smoothing for robotic manipulators by constructing Composite Trajectory Spline

Aiming at smoothing discontinuous corners in robotic tool paths, a novel spline named “Composite Trajectory Spline”(CT-Spline) is proposed in this paper. Based on CT-Spline, the real-time position path smoothing for robotic manipulators is realized. CT-Spline enables analytical trajectory planning and interpolation without curve length computation, significantly enhancing trajectory generation efficiency and avoiding numerical errors. By directly controlling resultant trajectory acceleration, CT-Spline avoids decomposing it into tangential and normal accelerations, ensuring adherence to the maximum acceleration constraint while fully utilizing acceleration potential. The geometric shape of CT-Spline is determined by three control points and a trajectory model. This paper further develops C1 and C3 continuous CT-Splines based on two different trajectory models, combined with a velocity look-ahead algorithm to achieve real-time local smoothing of path corners. Additionally, an adaptive control algorithm of smoothing error is introduced to dynamically maximize trajectory velocity. Simulations and experiments validate the effectiveness of CT-Spline, and demonstrate that the proposed C3 continuous trajectory smoothing method has several advantages over the method based on Pythagorean-hodograph splines in terms of kinematic constraints, velocity, smoothing errors, and real-time performance.

Path Planning and Motion Control of Robot Dog Through Rough Terrain Based on Vision Navigation.

This article delineates the enhancement of an autonomous navigation and obstacle avoidance system for a quadruped robot dog. Part one of this paper presents the integration of a sophisticated multi-level dynamic control framework, utilizing Model Predictive Control (MPC) and Whole-Body Control (WBC) from MIT Cheetah. The system employs an Intel RealSense D435i depth camera for depth vision-based navigation, which enables high-fidelity 3D environmental mapping and real-time path planning. A significant innovation is the customization of the EGO-Planner to optimize trajectory planning in dynamically changing terrains, coupled with the implementation of a multi-body dynamics model that significantly improves the robot's stability and maneuverability across various surfaces. The experimental results show that the RGB-D system exhibits superior velocity stability and trajectory accuracy to the SLAM system, with a 20% reduction in the cumulative velocity error and a 10% improvement in path tracking precision. The experimental results also show that the RGB-D system achieves smoother navigation, requiring 15% fewer iterations for path planning, and a 30% faster success rate recovery in challenging environments. The successful application of these technologies in simulated urban disaster scenarios suggests promising future applications in emergency response and complex urban environments. Part two of this paper presents the development of a robust path planning algorithm for a robot dog on a rough terrain based on attached binocular vision navigation. We use a commercial-of-the-shelf (COTS) robot dog. An optical CCD binocular vision dynamic tracking system is used to provide environment information. Likewise, the pose and posture of the robot dog are obtained from the robot's own sensors, and a kinematics model is established. Then, a binocular vision tracking method is developed to determine the optimal path, provide a proposal (commands to actuators) of the position and posture of the bionic robot, and achieve stable motion on tough terrains. The terrain is assumed to be a gentle uneven terrain to begin with and subsequently proceeds to a more rough surface. This work consists of four steps: (1) pose and position data are acquired from the robot dog's own inertial sensors, (2) terrain and environment information is input from onboard cameras, (3) information is fused (integrated), and (4) path planning and motion control proposals are made. Ultimately, this work provides a robust framework for future developments in the vision-based navigation and control of quadruped robots, offering potential solutions for navigating complex and dynamic terrains.

Feedback switching control of a walking piezoelectric actuator for trajectory tracking

Abstract This paper proposes a feedback switching control (FSC) scheme for a Walking Piezoelectric Actuator (WPA) to track a time-varying trajectory. A WPA has two feet with each foot integrated with a clamp and shear Piezoelectric Stack Actuators (PSAs). The first foot pushes the stage translator a step forward when clamped while the second foot is reset and retracted; subsequently, ‘foot switching’ occurs; as a result, the second foot is clamped to push the translator to make a next step forward and the first foot is reset and retracted. To make sure that the translator is always controlled in closed-loop no matter which foot is clamped, the FSC scheme is proposed. It consists of a clamping waveform generator which converts the trajectory into control signals for the clamp PSAs of the two feet and a feedback controller which provides control signals for the two shear PSAs of the feet. Experimental results show that the proposed control scheme achieves a tracking error of 0.028 µm for a constant velocity trajectory, which is a 75% reduction compared to the 0.113 µm of waveform-based control with ILC compensation in literature.

Numerical simulation and experimental analysis of aerosol deposition: Investigating nozzle effects on particle dynamics and deposition

Aerosol deposition (AD), also known as vacuum kinetic spray (VKS), is used to deposit dense ceramic films onto a substrate surface at room temperature. One predominant factor influencing the overall deposition process is the nozzle geometry, which significantly impacts the quality, shape, and thickness of the coating. To evaluate the effect of nozzle geometry in AD, particle trajectory and impact velocity were investigated via computational simulation. A Eulerian-Lagrangian model was used to simulate the gas flow, particle in-flight behavior, as well as particle deposition characteristics on a flat substrate. Three common nozzle geometries in AD were utilized: a converging-diverging nozzle with a slit cross-section (CD-Slit), a converging-barrel nozzle with a slit cross-section (CB-Slit), and a converging-diverging round (CD-Round) nozzle. Moreover, suitable drag coefficient and Nusselt number correlations were used to account for compressibility and rarefaction effects on particle dynamics and heat transfer. The simulation results were compared to the deposition pattern obtained experimentally. The results demonstrate that shape of the nozzle has profound effect on the particle impact velocity and deposition pattern. The CD-Round nozzle provides uniform, thinner coatings ideal for extensive surfaces at a slower deposition rate. In contrast, the CB-Slit nozzle is optimized for maximum coating thickness in narrow, thick linear patterns. The CD-Slit nozzle achieves high deposition rates with uniform coatings and a distinctive cat-ear shaped pattern.

A low-complexity solution for LMIs of hardly constrained nonlinear MIMO tracking control systems

This research is dedicated to the discrete-time tracking control of nonlinear multi-input multi-output (MIMO) Newtonian mechanics systems with an equal number of inputs and outputs, in the presence of input saturation and high-order dynamics of the actuation system. Firstly, the continuous-time model of the plant and actuation system has been transformed into the discrete-time domain using Adams–Bashforth and Adams–Moulton discretization techniques. In the next step, considering saturation boundaries, the modified position and velocity references are derived using the discrete-time reaching law-based control approach. These modified references are obtained through an optimization algorithm that calculates the closest achievable position and velocity trajectories with respect to the desired reference trajectory. Modifying desired trajectories in MIMO systems results in the formation of a system of linear matrix inequalities (LMIs). Solving such systems of inequalities requires the use of complex methods, and this difficulty increases with the increase in the system’s degrees of freedom (DOFs). The proposed strategy offers a more straightforward approach to solving these inequalities. Once the modified references are calculated, the corresponding position and velocity control signals can be obtained using the mentioned reaching law strategies. The final step of the proposed algorithm is to combine the resulting position and velocity control signals using a Gaussian-weighted averaging method to form the main control signal. A desirable performance for the control system can be achieved by appropriately adjusting the standard deviations of the Gaussian weighing functions. To evaluate the performance characteristics of the proposed method, it has been applied in the tracking control procedure of a nonlinear 3-DOF mechanical system. The numerical simulation results demonstrate the effectiveness of the method.

Sex-specific proximal tubular cell differentiation pathways identified by single-nucleus RNA sequencing

Postnatal kidney growth is substantial and involves expansion in kidney tubules without growth of new nephrons, which are the functional units of the kidney. Proliferation and differentiation pathways underpinning nephron elongation are not well defined. To address this, we performed sequential characterization of mouse kidney transcriptomics at the single cell level. Single nuclear RNA sequencing (snRNA-seq) was performed on kidney tissue from male and female mice at 1, 2, 4 and 12 weeks of age using the 10x Chromium platform. Unbiased clustering was performed on 68,775 nuclei from 16 animals. 31 discrete cellular clusters were seen, which were identified through comparison of their gene expression profiles to canonical markers of kidney cell populations. High levels of proliferation were evident at early time points in some cell types, especially tubular cells, but not in other cell types, for example podocytes. Proliferation was especially evident in Proximal Tubular Cells (PTCs) which are the most abundant cell type in the adult kidney. Uniquely when compared to other kidney cell types, PTCs demonstrated sex-specific expression profiles at late, but not early, time points. Mapping of PTC differentiation pathways using techniques including trajectory and RNA Velocity analyses delineated increasing PTC specialization and sex-specific phenotype specification. Our single-cell transcriptomics data characterise cellular states observed during kidney growth. We have identified PTC differentiation pathways that lead to sex-specific tubular cell phenotypes. Tubular proliferative responses are of central importance in postnatal kidney growth and have also been linked to kidney recovery versus fibrosis following injury. Our unbiased and comprehensive dataset of tubular cell development can be used to identify candidate pathways for therapeutic targeting.

Temporal adaptation of the postural control following a prolonged fin swimming.

Postural control deteriorates following a transition between two environments, highlighting a sensory conflict when returning to natural conditions. Aquatic immersion offers new perspectives for studying postural control adaptation in transitional situations. Our aim is to study immediate and post-task static postural control adaptation on land after a prolonged fin swimming exercise in total immersion. Standing static postural control was assessed in 14 professional or recreational SCUBA divers (11 men, 3 women; 33.21 ± 10.70years), with eyes open and closed, before, immediately after, and in the following 20min following a fully-immersed 45-min fin swimming exercise. Centre-of-pressure metrics (COP) including average position, amplitude, velocity, length and 95% ellipse were evaluated in medial-lateral (x-axis) and anterior-posterior (y-axis) directions with a force platform. The Romberg ratio was also assessed for each metric. A two-way repeated measures analysis of variance revealed a significant effect of the measurement period on COPx vel (p = 0.01), COPy vel (p < 0.01) and Length (p < 0.01), and of the visual condition on COPy vel (p < 0.01) and Length (p < 0.01). Eyes closed measures were systematically higher than eyes open measures despite there being no significant difference in the Romberg ratio in all periods. Post-immersion, the velocity and total trajectory of the centre of pressure remained systematically lower than baseline values in both visual conditions. Post-immersion, COP velocity and length significantly decreased, suggesting a sensory reweighting strategy potentially associated with ankle stiffening.

Probing the effect of time conformal factor on the particle dynamics around AdS Reissner–Nordström black hole

This study entails the effect of time on particle dynamics (both charged and uncharged) in the surrounding of the Reissner–Nordström AdS black hole in light of a general time conformal factor by acquiring Noether symmetry relation using the Lagrangian of the considered black hole spacetime. Upon acquiring the required time conformal spacetime as well as the Noether and approximated Noether symmetries along with their respective laws of conservation, the study further aims at investigating different parameters, like effective potential, effective force, and trajectories of escape velocity for the particle dynamics of the specified black hole. Additionally, in order to investigate orbit stability, the study employs the Lyapunov exponent.

Al-based amorphous coatings by warm spraying: Numerical simulation and experimental validation

Al-based fully amorphous coatings with low porosity were successfully produced via warm spraying, combining numerical simulation and experimental validation. A warm spray gun model, employing computational fluid dynamics, was developed to simulate the warm spraying process. The study delved into the effects of reactant flow rate, oxygen/fuel (O/F) ratio, coolant flow rate, spray distance, particle size, particle shape, particle injection velocity, and particle injection angle on the gas flow field parameters (pressure, temperature, and velocity) and particle in-flight behaviors (temperature, velocity, and in-flight trajectory). The optimum spraying parameters (OSP) predicted by the simulation results were determined: 0.008740 kg/s for reactant flow rate, 2.7 for the O/F ratio, 0.007356 kg/s for coolant flow rate, 142 mm for spraying distance, 10–35 μm for particle size range, spherical for particle shape, 10 m/s for particle injection velocity, and 0° for particle injection angle. Subsequently, warm spraying experiments based on the OSP yielded Al-based fully amorphous coatings with a porosity of 0.08 %. This work provides valuable insights for the production of Al-based fully amorphous coatings with minimal porosity through warm spraying.

The Influence of Reduced Frequency on H-VAWT Aerodynamic Performance and Flow Field Near Blades

Studies demonstrate that the reduced frequency k is influenced by the incoming wind speed U0 and the rotor speed n. As a dimensionless parameter, k characterizes the stability of the flow field, which is a critical factor affecting the performance of vertical-axis wind turbines (VAWTs). This paper investigates the impact of k on the performance of straight-blade vertical-axis wind turbines (H-VAWT). The findings indicate that 0.05 is the critical value of k. The same k results in a similar flow field structure, yet the performance changes vary with different U0. A decrease in n or an increase in U0 leads to an increase in the average value and fluctuation of k, which subsequently reduces the rotor rotation torque Cm and decreases the maximum wind energy utilization rate Cpmax. This reduction in Cpmax weakens the stability of the flow field. Additionally, the high-speed area of the blade’s trailing edge velocity trajectory at θ=0°, θ=120°, and θ=240° expands with increasing range. Velocity dissipation in the high-speed area of the trailing edge affects the stability of the flow field within the rotor.

PARALLEL PLATFORM CONTROLLER BASED ON ADAPTIVE DIFFERENCE ALGORITHM – PART 2

There are two main approaches to motion control on parallel platforms: joint space control and workspace control. Joint space control is an easy-to-implement semi-closed-loop strategy, but its control effect could be better. The workspace control is to obtain the real-time position of the parallel platform through the forward solution and close the speed and position loop of the parallel platform in the workspace. This paper uses a Model Predictive Controller (MPC) to control the parallel platform with workspace control as the research goal. The loss function is constructed based on the swarm intelligence optimization idea, and the adaptive difference algorithm is used to optimize the parameters of MPC. This part uses MATLAB to perform simulation experiments to complete the S-shaped velocity trajectory planning algorithm. In addition, the control effect of MPC and position-loop PI controller in a robust disturbance environment is compared. Experiments show that MPC has the advantages of low energy consumption and high control accuracy.

A modified bond-based peridynamic approach for rigid projectile perforation on concrete slabs

Peridynamic (PD) has a unique advantage in describing the crack growth and fragmentation of brittle materials. Concerning the dynamic behaviors and failure patterns of concrete slabs under projectile perforations, a modified bond-based PD approach maintaining both the easy implementation and computational stability characteristics was firstly developed from the following three aspects, (i) a rate-dependent PD constitutive model was proposed for describing the dynamic behaviors of concrete; (ii) a progressive damage criterion considering the tension-compression anisotropy, softening behavior, and strain rate effect of concrete was incorporated to more accurately reproduce the damage and failure of concrete; (iii) an improved micro-modulus function related to bond length was introduced to reveal the internal length effect of bond force. Then, numerical simulations of projectile perforation on concrete slabs by utilizing the developed modified bond-based PD approach, as well as the corresponding sensitivity analyses of discretization parameters including horizon size and particle spacing were performed. Based on the recommended horizon size and particle spacing, the predicted residual velocity of projectile and failure patterns of concrete slabs exhibited an excellent agreement with the test data. Furthermore, by comparisons of the traditional bond-based PD and classical finite element methods, the superiority of developed approach in describing the perforation damage of concrete targets against projectile impact was demonstrated. Finally, the modified bond-based PD approach was employed to blind simulate the projectile normal and oblique perforating multi-layered spaced concrete target plates. It was found that the modified PD model reasonably predicted the terminal ballistic trajectory, deflection angle, and residual velocity of projectile, as well as the failure patterns of target plates. The present work provides a new way to predict the terminal ballistic effect of projectile and dynamic behaviors of concrete slabs.

Single-cell sequencing reveals glial cell involvement in development of neuropathic pain via myelin sheath lesion formation in the spinal cord

BackgroundNeuropathic pain (NP), which results from injury or lesion of the somatosensory nervous system, is intimately associated with glial cells. The roles of microglia and astrocytes in NP have been broadly described, while studies on oligodendrocytes have largely focused on axonal myelination. The mechanisms of oligodendrocytes and their interactions with other glial cells in NP development remain uncertain.MethodsTo explore the function of the interaction of the three glial cells and their interactions on myelin development in NP, we evaluated changes in NP and myelin morphology after a chronic constriction injury (CCI) model in mice, and used single-cell sequencing to reveal the subpopulations characteristics of oligodendrocytes, microglia, and astrocytes in the spinal cord tissues, as well as their relationship with myelin lesions; the proliferation and differentiation trajectories of oligodendrocyte subpopulations were also revealed using pseudotime cell trajectory and RNA velocity analysis. In addition, we identified chemokine ligand-receptor pairs between glial cells by cellular communication and verified them using immunofluorescence.ResultsOur study showed that NP peaked on day 7 after CCI in mice, a time at which myelin lesions were present in both the spinal cord and sciatic nerve. Oligodendrocytes, microglia, and astrocytes subpopulations in spinal cord tissue were heterogeneous after CCI and all were involved in suppressing the process of immune defense and myelin production. In addition, the differentiation trajectory of oligodendrocytes involved a unidirectional lattice process of OPC-1-Oligo-9, which was arrested at the Oligo-2 stage under the influence of microglia and astrocytes. And the CADM1-CADM1, NRP1-VEGFA interactions between glial cells are enhanced after CCI and they had a key role in myelin lesions and demyelination.ConclusionsOur study reveals the close relationship between the differentiation block of oligodendrocytes after CCI and their interaction with microglia and astrocytes-mediated myelin lesions and NP. CADM1/CADM1 and NRP-1/VEGFA may serve as potential therapeutic targets for use in the treatment of NP.

Impacts of viscosity on bending behavior of the electrospun jet: Simulation model and experiment

Numerous experimental data indicate that higher viscosity solutions make it easier to obtain coarser fibers. However, the influences of viscosity on the whipping behavior of the jet and their subsequent impacts on the formation of terminal fiber during electrospinning remain topics of ongoing research. Simulation model and experiment presented in this study aim to solve these problems. The simulation model alters the dimensionless parameter Fve to predict the impacts of different viscosities on the bending behavior of the electrospun jet, while the experiment observes this behavior under different viscosities by regulating the solution concentration. The consistency of the simulation and experimental data implies the accuracy of the conclusions in this paper. Furthermore, the simulation model and experiment have demonstrated that a higher viscosity jet results in a smaller radius of the jet trajectory and lower jet velocity, which in turn leads to the formation of coarser fiber.

Growth pattern, growth deceleration, and relevant predictors in girls treated with GnRHa: a retrospective longitudinal study.

This study aimed to analyze the height growth pattern and the incidence of significant growth deceleration in girls with CPP and EFP on GnRHa treatment, and thereby identify relevant predictors of growth deceleration. The data of 99 girls diagnosed with CPP and 47 girls with EFP were included in this retrospective analysis. The incidence of growth deceleration was calculated in both the first and second years. Multivariate logistic regression analysis was used to identify predictors indicative of growth deceleration. Growth velocity (GV) trajectories showed gradual decreases to the nadir at 18months of treatment, and then they recovered till the 24th month of treatment, especially in girls with CPP. Nevertheless, the recovery was significantly greater in the CPP group than EFP. In the first year, no significant difference in the incidence of growth deceleration was found between the CPP group and the EFP group [17.35 vs. 25.53 %, p=0.249]; in the second year, the CPP group had alower incidence than the EFP group [42.86 vs. 76.92 %, p=0.027]. The multivariate logistic regression analysis suggested that bone age (BA) was an independent predictor of growth deceleration (OR=2.264, 95 % CI: 1.268-4.042, p=0.006). The result of ROC curves showed the cut-off value of BA was 11.05 years. GV varies at different periods during GnRHa treatment. GnRHa should be used with more caution for EFPtreatment than for CPP. BA can be used to predict the occurrence of growth deceleration during GnRHa treatment.

A Robust Multi-Camera Vehicle Tracking Algorithm in Highway Scenarios Using Deep Learning

In intelligent traffic monitoring systems, the significant distance between cameras and their non-overlapping fields of view leads to several issues. These include incomplete tracking results from individual cameras, difficulty in matching targets across multiple cameras, and the complexity of inferring the global trajectory of a target. In response to the challenges above, a deep learning-based vehicle tracking algorithm called FairMOT-MCVT is proposed. This algorithm con-siders the vehicles’ characteristics as rigid targets from a roadside perspective. Firstly, a Block-Efficient module is designed to enhance the network’s ability to capture and characterize image features across different layers by integrating a multi-branch structure and depth-separable convolutions. Secondly, the Multi-scale Dilated Attention (MSDA) module is introduced to improve the feature extraction capability and computational efficiency by combining multi-scale feature fusion and attention mechanisms. Finally, a joint loss function is crafted to better distinguish between vehicles with similar appearances by combining the trajectory smoothing loss and velocity consistency loss, thereby considering both position and velocity continuity during the optimization process. The proposed method was evaluated on the public UA-DETRAC dataset, which comprises 1210 video sequences and over 140,000 frames captured under various weather and lighting conditions. The experimental results demonstrate that the FairMOT-MCVT algorithm significantly enhances multi-target tracking accuracy (MOTA) to 79.0, IDF1 to 84.5, and FPS to 29.03, surpassing the performance of previous algorithms. Additionally, this algorithm expands the detection range and reduces the deployment cost of roadside equipment, effectively meeting the practical application requirements.