Accurate Motion Prediction Research Articles

The CNN-LSTM-attention model for short term prediction of the polar motion

Abstract The accuracy of polar motion (PM) prediction significantly impacts the fields of coordinate frame transformation, satellite orbit determination, and deep space exploration. The present study develops two short term forecasting models based on the EOP 14C04 series. One hybrid approach incorporates convolutional neural networks (CNNs) and long short-term memory networks (LSTM), augmented with an attention mechanism; whereas another baseline model comprises CNN and LSTM. The first model, in contrast to the second model, incorporates an attention mechanism module for a more comprehensive integration of temporal information at each time step. In the initial short-term forecasting experiment, we conducted 360 repeated predictions, and the findings revealed that the parameters suitable for PMX forecasting may not necessarily be applicable to PMY forecasting. In the second experiment, the two models generated a total of 500 forecasts, each encompassing short-term predictions ranging from 1 to 30 d. The experimental results demonstrate that the first model exhibits mean absolute error (MAE) range of 0–7.72 mas for PMX and 0–4.73 mas for PMY, while the second model shows MAE range of 0–7.88 mas for PMX and 0–4.78 mas for PMY. After two exploratory experiments, we discovered the following results: the first model exhibits marginally superior predictive accuracy compared to the second model. Furthermore, this study substantiates the robustness of both models in short-term prediction and affirms the significance of assigning distinct weights to past temporal intervals in forecasting, thereby offering a novel perspective for PM prediction research.

Design and validation of finite element models for the assessment of post-fixation distal femur fracture motion.

Fracture site motion is thought to play an important role in the healing of complex fractures of the distal femur via mechanotransduction. Measuring this motion in vivo is challenging, and this has led researchers to turn to finite element modeling approaches to gain insights into the mechanical environment at the fracture site. Developing a systematic understanding of the effect of different model choices for distal femur fractures may allow more accurate prediction of fracture site motion from these types of simulations. In this study, we aim to assess the effect of four different modeling choices and parameters. We looked at the effect of using bone specific density distributions vs generic values, employing landmark-based geometry generation, varying fracture alignment within clinically relevant ranges, and determining whether direct apposition of the fracture to the plate was achieved. For validation, five cadaveric femurs had fractures created and repaired with plated constructs, and these were then loaded and fracture site motion was directly measured. We found that using landmark based bone geometry and patient-specific bone density distributions had a minimal effect on the overall model predictions. Changing the alignment, particularly into varus and procurvatum could have a large (>50%) effect on predicted shear motion, as could direct apposition of the bone to the plate. These findings demonstrate that modeling choices can play an important role in simulating distal femur fracture mechanics, and it is particularly critical that patient customized models attempt to accurately represent alignment of the bone fragments and lateral plate apposition.

Numerical simulation and analysis of motion responses and added resistance of a KRISO container ship in regular waves

Abstract In order to explore the influence of vessel speed on seakeeping characteristics and provide valuable insights for ship performance evaluation, the accurate prediction of ship motion and resistance in wave conditions is essential. In this study, the variations in ship motion and resistance with respect to ship speed are analyzed during navigation in regular waves. The results suggest that the added resistance coefficient of the KCS demonstrates an initial rise, succeeded by a subsequent decline, with the expansion of wavelength. Moreover, the peak non-dimensional resistance coefficient, observed at different ship speeds, demonstrates an upward trend with the augmentation of ship velocity. This research methodology provides a valuable tool for numerically predicting the hydrodynamic performance of vessels in regular wave conditions.

Leveraging the Empirical Wavelet Transform in Combination with Convolutional LSTM Neural Networks to Enhance the Accuracy of Polar Motion Prediction

High-precision polar motion prediction is of great significance for deep space exploration and satellite navigation. Polar motion is affected by a variety of excitation factors, and nonlinear prediction methods are more suitable for polar motion prediction. In order to explore the effect of deep learning in polar motion prediction. This paper proposes a combined model based on empirical wavelet transform (EWT), Convolutional Neural Networks (CNN) and Long Short Term Memory (LSTM). By training and forecasting EOP 20C04 data, the effectiveness of the algorithm is verified, and the performance of two forecasting strategies in deep learning for polar motion prediction is explored. The results indicate that recursive multi-step prediction performs better than direct multi-step prediction for short-term forecasts within 15 days, while direct multi-step prediction is more suitable for medium and long-term forecasts. In the 365 days forecast, the mean absolute error of EWT-CNN-LSTM in the X direction and Y direction is 18.25 mas and 15.78 mas, respectively, which is 23.5% and 16.2% higher than the accuracy of Bulletin A. The results show that the algorithm has a good effect in medium and long term polar motion prediction.

Multi-granularity spatial temporal graph convolution network with consecutive attention for human motion prediction

Human motion prediction is attracting increasing attention for its numerous potential applications in fields including autonomous driving, video surveillance and virtual reality. However, accurate motion prediction is challenging due to the complex spatial dependencies, dynamic temporal correlations and high dimension of human pose sequences. Existing graph-based methods rarely consider positional and channel information of the feature map, resulting in lower prediction accuracy. Therefore, we propose a novel multi-granularity spatial temporal graph convolution network with consecutive attention (MSTCA) for human motion prediction. Firstly, a multi-granularity spatial convolution network is introduced to capture spatial joint features through multiple kernel sizes. Then, consecutive attention module is proposed to capture both positional and channel information of the feature map. Next, MSTCA uses multi-granularity temporal convolutional network to extract temporal correlations with multiple receptive fields and predict future poses. Finally, a decoder composed of a 2D convolution layer and several PRelu layers integrates the output of the whole model. Experimental results on the GTA-IM and PROX datasets demonstrate that our method significantly improves the accuracy of human motion prediction in comparison to the existing approaches.

A new wake superposition model and its application in collision prevention control of tri-riser array system

In the field of offshore oil exploitation, the compact multi-riser arrangement increases the risk of collision and has an adverse impact on production and the marine environment. The solution to this problem involves accurate motion prediction and collision prevention control under the effect of flow field superposition between risers, but there is little research on this related theory. This paper improves the traditional wake shielding effect model and further proposes a wake shielding superposition effect model for the mutual influence between risers, in order to establish a comprehensive accurate motion control model for tri-riser array system. A solution method is introduced for this model, both vibration and collision analysis are conducted for the system. The study revealed that solely augmenting the spacing inadequately mitigates system collisions, and active control is an inevitable choice. Based on this, this paper further designs an active boundary controller that is easy to implement in engineering, to achieve collaborative collision prevention control for three risers, and verifies its stability. Compared with traditional PD control, the proposed control method exhibits significant superiority, which can effectively reduce vibration amplitude, significantly increase dynamic spacing, and thus effectively prevent collisions between risers.

Real-time prediction of ship motions based on the reservoir computing model

Real-time prediction of ship motions is crucial for ensuring the safety of offshore activities. In this study, we investigate the performance of the reservoir computing (RC) model in predicting the motions of a ship sailing in irregular waves, comparing it with the long short-term memory (LSTM), bidirectional LSTM (BiLSTM), and gated recurrent unit (GRU) networks. The model tests are carried out in a towing tank to generate the datasets for training and testing the machine learning models. First, we explore the performance of machine learning models trained solely on motion data. It is found that the RC model outperforms the LSTM, BiLSTM, and GRU networks in both accuracy and efficiency for predicting ship motions. Besides, we investigate the performance of the RC model trained using the historical motion and wave elevation data. It is shown that, compared with the RC model trained solely on motion data, the RC model trained on the motion and wave elevation data can significantly improve the motion prediction accuracy. This study validates the effectiveness and efficiency of the RC model in ship motion prediction during sailing and highlights the utility of wave elevation data in enhancing the RC model’s prediction accuracy.

Deck motion prediction using neural kernel network Gaussian process regression

Deck motion prediction and compensation are critical technologies for carrier-based aircraft automatic landing. Traditional deck motion prediction methods rely on precision of motion models and parameter adjustments, facing challenges in adaptability to complex sea conditions, different types of carriers, changes in flight conditions, and limitations in prediction duration, as well as reliability issues. This paper proposes a deck motion prediction method based on the neural kernel network Gaussian process regression (NKN-GPR) model. The NKN-GPR model can utilize a neural kernel network (NKN) to automatically construct the Gaussian process regression (GPR) model′s composite kernel, effectively addressing the limitations of the automated kernel search (ACKS) algorithm, which heavily depends on manual prior knowledge. Simulation data is generated using a combination of sine wave and power spectrum models, and the NKN-GPR model is compared with an autoregressive (AR) model based on least squares in a simulated validation. The simulation results demonstrate that the NKN-GPR model exhibits significant advantages in motion prediction accuracy, smoothness, and prediction duration, which confirms the effectiveness of the proposed algorithm. This study provides theoretical support for safe automatic landing of carrier-based aircraft.

Explicitly encoding the cyclic nature of breathing signal allows for accurate breathing motion prediction in radiotherapy with minimal training data

Background and purposeActive breathing motion management in radiotherapy consists of motion monitoring, quantification and mitigation. It is impacted by associated latencies of a few 100 ms. Artificial neural networks can successfully predict breathing motion and eliminate latencies. However, they require usually a large dataset for training. The objective of this work was to demonstrate that explicitly encoding the cyclic nature of the breathing signal into the training data enables significant reduction of training datasets which can be obtained from healthy volunteers. Material and methodsSeventy surface scanner breathing signals from 25 healthy volunteers in anterior-posterior direction were used for training and validation (ratio 4:1) of long short-term memory models. The model performance was compared to a model using decomposition into phase, amplitude and a time-dependent baseline. Testing of the models was performed on 55 independent breathing signals in anterior-posterior direction from surface scanner (35 lung, 20 liver) of 30 patients with a mean breathing amplitude of (5.9 ± 6.7)mm. ResultsUsing the decomposed breathing signal allowed for a reduction of the absolute root-mean square error (RMSE) from 0.34 mm to 0.12 mm during validation. Testing using patient data yielded an average absolute RMSE of the breathing signal of (0.16 ± 0.11)mm with a prediction horizon of 500 ms. ConclusionIt was demonstrated that a motion prediction model can be trained with less than 100 datasets of healthy volunteers if breathing cycle parameters are considered. Applied to 55 patients, the model predicted breathing motion with a high accuracy.

A review on motion sickness of autonomous driving vehicles

The objective of this study is to investigate the symptoms, types, etiology, and assessment methods of motion sickness in autonomous vehicles in order to gain a comprehensive understanding of its occurrence mechanism and emphasize the significance of enhancing autonomous vehicle algorithms for improved ride comfort. Thus, this paper provides a synthesis and discussion of various theories while exploring strategies for mitigating motion sickness from three perspectives: passengers, vehicles, and external equipment. Firstly, it summarizes the clinical manifestations and classification of motion sickness while conducting an in-depth analysis of associated factors. Secondly, it evaluates different approaches for quantitatively measuring the severity and extent of motion sickness. Subsequently, it analyzes the reasons behind increased motion sickness caused by autonomous vehicles and emphasizes the importance of algorithmic improvements to enhance travel comfort. Finally, mitigation strategies are proposed considering passengers' needs as well as advancements in accurate motion prediction models and optimization techniques for autonomous planning and control algorithms that can effectively reduce the risk of motion sickness. As application scenarios for autonomous technology continue to expand, meeting user requirements while ensuring safety has become a benchmark for assessing technical proficiency. Therefore, promoting unmanned travel services necessitates a thorough analysis of existing issues related to autonomous technology along with prioritizing algorithm design enhancements through effective means to achieve an enhanced user experience.

Multivariate USV Motion Prediction Method Based on a Temporal Attention Weighted TCN-Bi-LSTM Model

Unmanned surface vehicle (USV)’s motion is represented by time-series data that exhibit highly nonlinear and non-stationary features, significantly influenced by environmental factors, such as wind speed and waves, when sailing on the sea. The accurate prediction of USV motion, particularly crucial parameters, such as the roll angle and pitch angle, is imperative for ensuring safe navigation. However, traditional and single prediction models often struggle with low accuracy and fail to capture the intricate spatial–temporal dependencies among multiple input variables. To address these limitations, this paper proposes a prediction approach integrating temporal convolutional network (TCN) and bi-directional long short-term memory network (Bi-LSTM) models, augmented with a temporal pattern attention (TPA) mechanism, termed the TCN-Bi-LSTM-TPA (TBT) USV motion predictor. This hybrid model effectively combines the strengths of TCN and Bi-LSTM architectures to extract long-term temporal features and bi-directional dependencies. The introduction of the TPA mechanism enhances the model’s capability to extract spatial information, crucial for understanding the intricate interplay of various motion data. By integrating the features extracted by TCN with the output of the attention mechanism, the model incorporates additional contextual information, thereby improving prediction accuracy. To evaluate the performance of the proposed model, we conducted experiments using real USV motion data and calculated four evaluation metrics: mean square error (MSE), mean absolute error (MAE), mean absolute percentage error (MAPE), and R-squared (R2). The results demonstrate the superior accuracy of the TCN-Bi-LSTM-TPA hybrid model in predicting USV roll angle and pitch angle, validating its effectiveness in addressing the challenges of multivariate USV motion prediction.

Simplified neural architecture for efficient human motion prediction in human-robot interaction

Human motion prediction is essential for safe and effective human-robot interaction, but modeling the intricate spatio-temporal dynamics inherent in human movement remains challenging. While recent methods have advanced motion prediction accuracy, they rely on complex neural architectures such as Graph Convolutional Networks and Recurrent Neural Networks, demanding extensive hyperparameter tuning. To overcome this limitation, we propose a streamlined multi-stage layer incremental perceptron (MSLP) architecture that achieves competitive results with far fewer parameters, improving efficiency. The MSLP eschews the complexity of prevalent networks and instead takes a stepped refinement approach to predict intricate motions using a lightweight model. This simplified yet effective design enables nuanced learning of spatio-temporal relationships without exhaustive tuning. Specifically, the MSLP incorporates two key components: a multi-channel feature extraction and enhancement block (MFE-block) and a temporal feature extraction module (TFE-block). The MFE-block strengthens the representation of each action node by integrating multi-dimensional action features. The TFE-block then captures the contextual relationships between actions over time. Together, the MFE-block and TFE-block allow the MSLP to model the complex spatial and temporal dynamics of human movement using a streamlined architecture and minimal parameter tuning. When evaluated on established datasets including Human3.6 M, CMU, 3DPW, and AMASS, the proposed MSLP method achieves improved accuracy gains of 16.7%-67.0% over existing state-of-the-art techniques. Additionally, the MSLP significantly reduces the number of parameters by 35.7%-99.5% compared to prior architectures. We find that the proposed MSLP significantly improves both long-term prediction and generalization capabilities of the model.

Short-Term Prediction of Ship Roll Motion in Waves Based on Convolutional Neural Network

In this study, a short-term prediction method for ship roll motion in waves based on convolutional neural network (CNN) is presented. Firstly, based on the ship roll motion equation, the data for free roll attenuation motion in still water, roll motion in regular waves, and roll motion excited by irregular waves are simulated, respectively. Secondly, the simulation data is normalized and preprocessed, and then the time-sliding window technique is applied to construct the training and testing sample sets. Thirdly, the CNN model is trained by learning from the constructed training sample sets, and the well-trained CNN model is applied to predict the roll motion. To validate the CNN model’s prediction accuracy and effectiveness, a comparison between the forecasted results and the simulation data is conducted. Meanwhile, the predicted results are also compared with that of the long-short-term memory (LSTM) neural network. The research results demonstrate that CNN can effectively achieve accurate prediction of ship roll motion in waves, and its prediction accuracy is the same as that of the LSTM neural network.

A Motion Logic Network for Pedestrian Motion Prediction

Accurate and fast motion prediction such as pedestrian motion prediction (PMP) is crucial for safe autonomous driving. Much research effort has been devoted to studying the reactive behaviors of pedestrians, such as the interaction between pedestrians or the interaction between pedestrians and the environment. However, compared with behavioral logic, the current motion state of pedestrians has a greater influence on the future trajectory. In this work, we propose a motion logic network (MLN) to improve both the accuracy and efficiency of pedestrian motion prediction. Compared with the traditional data-driven neural networks, the concept of motion logic is directly introduced into the network so that the training of the network is not required. In particular, instead of fitting the network based on inputs and target outputs, the proposed method directly adopts motion logic to predict the future trajectory. To illustrate the performance of the proposed MLN, several experiments have been performed. For acceleration and deceleration motion in practical experiment, the average displacement error (ADE) of MLN has an improvement of as high as 6.25% than CVM’s, while the final displacement error (FDE) of MLN has an improvement of 11.1%. In terms of efficiency, MLN is a hundred times faster than LSTM. It indicates that motion logic plays an important role in the development of prediction algorithms for pedestrian motion. <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Note to Practitioners</i> —This article is motivated by the effects of pedestrians’ different motion states on the future trajectory. For example, the pedestrians’ state may also change drastically in some situations such as dashing across the road or stopping in anticipation of a bicycle crossing the path. This work explores the use of pedestrians’ physical motion states to develop a network that emphasizes the logical features of pedestrian motion so as to directly estimate the future motion trajectory and trend based on the motion logic. For the first time, the concept of motion logic and the resultant training-free motion logic network (MLN) is proposed, which takes into account the accuracy and efficiency performance of the algorithm. Compared with the state-of-the-art pedestrian prediction methods, the proposed method can better predict the trajectory of pedestrians in more complex motion states. Moreover, a series of simulations and practical experiments with pseudo-constant velocity motion and acceleration/deceleration motion was taken to verify the performance of the proposed MLN.

A high-dimensional respiratory motion modeling method based on machine learning

PurposeRespiratory motion of the human body remains a major challenge in procedures for percutaneous puncture intervention. A solution is to develop a respiratory motion model. The aim of this study is to construct a respiratory motion model that can accommodate extreme respiratory conditions. MethodsPrincipal component analysis (PCA) and the support vector regression (SVR) method were employed as the framework to build the respiratory motion model. In the first stage, internal respiratory signals and respiratory surrogate signals are extracted. In the second stage, the respiratory motion model based on SVR is established. This study proposes a unique data augmentation method to improve the model robustness and the corresponding response capacity under extreme respiratory conditions. Finally, relevant evaluation indicators are used to evaluate the motion prediction accuracy of a reference model and the model in this study. ResultsWhen estimating the motion of the end-inspiratory phase (T00), the mean target registration error (TRE) of the motion estimate of the model after data augmentation was 1.28 ± 0.24 mm, while that of the model before data augmentation and that of the reference model were 1.67 ± 0.55 mm and 2.29 ± 1.06 mm, respectively. When estimating the motion of all phases, the mean TRE of the model after data augmentation was 1.26 ± 0.23 mm, which was also lower than that of the model before data augmentation. The data augmentation method improves the accuracy of the model and is statistically significant. ConclusionDisplaying excellent performance under extrapolation conditions, the proposed model effectively improves both accuracy and robustness.

Improving the accuracy of polar motion prediction using a hybrid least squares and long short-term memory model

Earth rotation parameters (ERPs) are essential for transforming between the celestial and terrestrial reference frames, and for high-precision space navigation and positioning. Among the ERPs, polar motion (PM) is a critical parameter for analyzing and understanding the dynamic interaction between the solid Earth, atmosphere, ocean, and other geophysical fluids. Traditional methods for predicting the change in ERPs rely heavily on linear models, such as the least squares (LS) and the autoregressive (AR) model (LS + AR). However, variations in ERP partly reflect non-linear effects in the Earth system, such that the predictive accuracy of linear models is not always optimal. In this paper, long short-term memory (LSTM), a non-linear neural network, is employed to improve the prediction of ERPs. Polar motion prediction experiments in this study are conducted using the LSTM model and a hybrid method LS + LSTM model based on the IERS EOP14C04 time series. Compared with Bulletin A, the PMX and PMY prediction accuracy can reach a maximum of 33.7% and 31.9%, respectively, with the LS + LSTM model. The experimental results show that the proposed hybrid model displays a better performance in mid- and long-term (120–365 days) prediction of polar motion.Graphical

Real Time 4D-Cone Beam CT via Single-Angle Projection with Dual Attention Mechanism

The hypothesis is that dual attention integrated Residual Network (DA-ResNet) together with a Principal Component Analysis (PCA) model accurately maps the measured on-treatment CBCT projections from different phases to each phase's 3D-CBCT real timely. Finally, it achieves real time 4D-CBCT. The DA-ResNet contains a CBAM (Convolutional Block Attention Module) and a SPA (Spatial Pyramid Structure). Hundreds of forward projections from one view of patient 4D-CT were simulated to train DA-ResNet. The extracted motion features were expressed as PCA coefficients. To test DA-ResNet performance, several single measured on-treatment projection is sent into it and the network output several groups of PCA coefficients real timely. Each group of PCA coefficient stands for the corresponding deformation field between the 0% phase and the predicted phase that the PCA label stands for. In labels such as Mean Absolute Error (MAE), R square, Normalized Cross Correlation (NCC), Structural Similarity Index Measure (SSIM) were used to evaluate DA-ResNet performance with other state-of-the-art networks such as CNN/Unet/ResNet. DA-ResNet outperforms other popular networks on all of the quantification labels, with MAE (14.41 ± 8.85), R square (0.95 ± 0.07) in PCA coefficient prediction, and NCC (0.9981 ± 0.0006), SSIM (0.98 ± 0.04) in image reconstruction. Compared with conventional ResNet, DA-ResNet reduced MAE by 79%. Meanwhile R square rises from 0.88 to 0.95, NCC rises from 0.996 to 0.998, SSIM rises from 0.94 to 0.98. DA-ResNet offers a solution for real time on-treatment 4D-CBCT image guidance by using single 2D on-line projections. It offers better motion prediction and image reconstruction accuracy compared with traditional deep learning networks (e.g., CNN/Unet/ResNet). The initial results are promising. The prospect of clinical application needs further clinical verification.

Probabilistic analysis of the instantaneous center of rotation of a floating structure under regular and irregular beam waves

Accurate prediction of roll motion is crucial to the safety of ships and floating structures. The “roll center” mentioned in the literature is ambiguous and various definitions exist. Hence, the instantaneous center of rotation (ICR) is preferred, being an established kinematic concept. Understanding the ICR is important on account of its significant influence on roll damping. A previous study postulated that the ICR moves in a straight line for regular beam waves, based on experimental observations. In this paper, the probability distribution of the ICR coordinates (y1, y2) is derived analytically for regular and irregular beam waves. For regular waves, it is proven a fortiori that the ICR lies on a straight line, and that y1 and y2 follow the Cauchy distribution, for which the mean and variance are undefined. For irregular waves, it is shown that the joint probability density of y1 and y2 is a bivariate Cauchy distribution, thus the iso-density contour lines are concentric ellipses. Regular and irregular wave experiments are performed to investigate realistic conditions. The analytical solutions are found to be quite accurate, notwithstanding some discrepancies arising from nonlinearities and experimental imperfections. The theoretical results presented herein have potential practical applications, for example since the ICR governs damping, the ICR distribution allows damping to be modeled as a stochastic process.

Sparse Bayesian Relevance Vector Machine Identification Modeling and Its Application to Ship Maneuvering Motion Prediction

In order to establish a sparse and accurate ship motion prediction model, a novel Bayesian probability prediction model based on relevance vector machine (RVM) was proposed for nonparametric modeling. The sparsity, effectiveness, and generalization of RVM were verified from two aspects: (1) the processed Sinc function dataset, and (2) the tank test dataset of the KRISO container ship (KCS) model. The KCS was taken as the main research plant, and the motion prediction models of KCS were obtained. The ε-support vector regression and υ-support vector regression were taken as the compared algorithms. The sparsity, effectiveness, and generalization of the three algorithms were analyzed. According to the trained prediction models of the three algorithms, the number of relevance vectors was compared with the number of support vectors. From the prediction results of the Sinc function and tank test datasets, the highest percentage of relevance vectors in the trained sample was below 17%. The final prediction results indicated that the proposed nonparametric models had good prediction performance. They could ensure good sparsity while ensuring high prediction accuracy. Compared with the SVR, the prediction accuracy can be improved by more than 14.04%, and the time consumption was also relatively lower. A training model with good sparsity can reduce prediction time. This is essential for the online prediction of ship motion.

GGTr: An Innovative Framework for Accurate and Realistic Human Motion Prediction

Human motion prediction involves forecasting future movements based on past observations, which is a complex task due to the inherent spatial-temporal dynamics of human motion. In this paper, we introduced a novel framework, GGTr, which adeptly encapsulates these patterns by integrating positional graph convolutional network (GCN) layers, gated recurrent unit (GRU) network layers, and transformer layers. The proposed model utilizes an enhanced GCN layer equipped with a positional representation to aggregate information from body joints more effectively. To address temporal dependencies, we strategically combined GRU and transformer layers, enabling the model to capture both local and global temporal dependencies across body joints. Through extensive experiments conducted on Human3.6M and CMU-MoCap datasets, we demonstrated the superior performance of our proposed model. Notably, our framework shows significant improvements in predicting long-term movements, outperforming state-of-the-art methods substantially.