Search

Copyright of Journal of South-Central Minzu University (Natural Science Edition) is the property of Journal of South-Central Minzu University (Natural Science Edition) Editorial Office and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Recently, human motion prediction has gained significant attention and achieved notable success. However, current methods primarily rely on training and testing with ideal datasets, overlooking the impact of variations in the viewing distance and viewing angle, which are commonly encountered in practical scenarios. In this study, we address the issue of model invariance by ensuring robust performance despite variations in view distances and angles. To achieve this, we employed Riemannian geometry methods to constrain the learning process of neural networks, enabling the prediction of invariances using a simple network. Furthermore, this enhances the application of motion prediction in various scenarios. Our framework uses Riemannian geometry to encode motion into a novel motion space to achieve prediction with an invariant viewing distance and angle using a simple network. Specifically, the specified path transport square-root velocity function is proposed to aid in removing the view-angle equivalence class and encode motion sequences into a flattened space. Motion coding by the geometry method linearizes the optimization problem in a non-flattened space and effectively extracts motion information, allowing the proposed method to achieve competitive performance using a simple network. Experimental results on Human 3.6M and CMU MoCap demonstrate that the proposed framework has competitive performance and invariance to the viewing distance and viewing angle. [ABSTRACT FROM AUTHOR]

The disturbance of the coal caving mechanism is the main method to solve the problem of coal falling into an arch during the top coal caving process, and the clearance in the caving mechanism is an important factor affecting its disturbance. To study the influence of clearance on the hydraulic supports, the dynamic analysis of hydraulic supports with two-level and three-level caving mechanisms is conducted considering the clearance. Based on Lagrange's equations of first kind, the mathematical model is constructed, and the equivalent spring stiffness of the oil cylinder is solved. The dynamic response of hydraulic support is studied by solving the mathematical model, and the relative position changes of the journal and bearing are compared. The research results prove that clearance has a better disturbance effect on the three-level caving mechanism. At the same time, the influence of clearance size and motion process on the motion error of the three-level caving mechanism is analyzed. Using the Kriging surrogate model and polynomial interpolation, the motion prediction model for the secondary tail beam with different clearance sizes is proposed, and the predicted values of the test samples are compared with the numerical solution values to prove the feasibility. [ABSTRACT FROM AUTHOR]

The visual naturalness of a rendered character's motion is an important factor in computer graphics work, and the rendering of jumping motions is no exception to this. However, the computational mechanism that underlies the observer's judgement of the naturalness of a jumping motion has not yet been fully elucidated. We hypothesized that observers would perceive a jumping motion as more natural when the jump trajectory was consistent with the trajectory of a vertical projectile motion based on Earth's gravity. We asked human participants to evaluate the naturalness of point-light jumping motions whose height and duration were modulated. The results showed that the observers' naturalness rating varied with the modulation ratios of the jump height and duration. Interestingly, the ratings were high even when the height and duration differed from the actual jump. To explain this tendency, we constructed computational models that predicted the theoretical trajectory of a jump based on the projectile motion formula and calculated the errors between the theoretical and observed trajectories. The pattern of the errors correlated closely with the participants' ratings. Our results suggest that observers judge the naturalness of observed jumping motion based on the error between observed and predicted jump trajectories. [ABSTRACT FROM AUTHOR]

In this study, we introduce a novel method for an online trajectory replanning approach for fixed-wing Unmanned Aerial Vehicles (UAVs). Our method integrates moving obstacle predictions within a gradient-based optimization framework. The trajectory is represented by uniformly discretized waypoints, which serve as the optimization variables within the cost function. This cost function incorporates multiple objectives, including obstacle avoidance, kinematic and dynamic feasibility, similarity to the reference trajectory, and trajectory smoothness. To enhance prediction accuracy, we combine physics-based and pattern-based methods for predicting obstacle movements. These predicted movements are then integrated into the online trajectory replanning framework, significantly enhancing the system's safety. Our approach provides a robust solution for navigating dynamic environments, ensuring both optimal and secure UAV operation. [ABSTRACT FROM AUTHOR]

Multi-object tracking (MOT) aims at estimating locations and identities of objects in videos. Many modern multiple-object tracking systems follow the tracking-by-detection paradigm, consisting of a detector followed by a method for associating detections into tracks. Tracking by associating detections through motion-based similarity heuristics is the basic way. Motion models aim at utilizing motion information to estimate future locations, playing an important role in enhancing the performance of association. Recently, a large-scale dataset, DanceTrack, where objects have uniform appearance and diverse motion patterns, was proposed. With existing hand-crafted motion models, it is hard to achieve decent results on DanceTrack because of the lack of prior knowledge. In this work, we present a motion-based algorithm named TLtrack, which adopts a hybrid strategy to make motion estimates based on confidence scores. For high confidence score detections, TLtrack employs transformers to predict its locations. For low confidence score detections, a simple linear model that estimates locations through trajectory historical information is used. TLtrack can not only consider the historical information of the trajectory, but also analyze the latest movements. Our experimental results on the DanceTrack dataset show that our method achieves the best performance compared with other motion models. [ABSTRACT FROM AUTHOR]

Fingerling counting is a basic operation in fish farming and provides an important guideline for many aspects of aquaculture. However, most of the current counting methods are inefficient or computationally cumbersome. This study proposed a high-speed, non-intrusive fingerling counting method based on multiple detectors and a motion prediction model, which achieved high-accuracy counting under the condition of low-frame rate. Firstly, to effectively detect and locate the adherent fingerlings, the detector was accomplished by combining the mixture of Gaussian-based (MOG) segmentation algorithm and the local extremum-based blob detection algorithm. Secondly, three different functions were used to construct a motion prediction model to predict the approximate probability of the fingerlings at each position in the previous frame. Thirdly, the cost matrix was constructed with probability as the feature to associate the fingerlings in the consecutive frames, and the newly appeared fingerlings were counted in real-time, realising the continuous fingerling counting with high precision. Through testing and analysis on 52 collected datasets under low-frame-rate (10 fps) acquisition conditions using largemouth bass (Micropterus salmoides , 3–5 cm) and crucian carp (Carassius auratus , 2–6 cm), results indicated that the best motion prediction model with segmentation function reached over 99% average counting accuracy for both species, with a standard deviation of accuracy less than 0.7%. This method provides a low-cost, high-speed, and stable application solution for computer vision-based fingerling counting. • Combined Mixture of Gaussian segmentation and blob detection for fingerling counting. • Integration of a motion prediction model achieved over 99% average counting accuracy. • Open access to a dataset containing annotation files for fingerlings counting. [ABSTRACT FROM AUTHOR]

Human motion prediction in various motion capture applications, e.g., optical and inertial, is challenging because of the complexity of human motion sequences. Current studies on this issue have insufficient analysis on the latent motion information in a given motion sequence, such as motion trends, transient changes, and temporal evolution. Meanwhile, methods using simple graph convolution networks suffer from over-smoothing, causing the predicted poses staying invariant in long-term prediction. To address these challenges, we propose a multi-feature-based orthogonal graph convolution network (MFOGCN), where the multi-feature extraction consists of two key modules: (1) hybrid spectral transform, which captures local transient features and global motion trends of motion sequences by discrete wavelet transform while considering temporal smoothing between human joints and (2) mask-aware multiple attention, with sliding time windows to extract motion sequence feature representations from historical multiple subsequences, refining the correlation between adjacent poses while obtaining global dependencies between sequences. In addition, we propose orthogonal graph convolution and orthogonal loss for the prediction network, which help to stabilize the feature transformation of the graph convolution to resolve the over-smoothing issue. An extensive evaluation on the Human 3.6M, AMASS and 3DPW datasets has been conducted, showing reliable effectiveness of the proposed MFOGCN that outperforms other approaches. [ABSTRACT FROM AUTHOR]

As an important interface between unmanned aerial vehicles (UAVs) and ships, the stability and motion control compensation technology of the shipborne UAV landing platform are paramount for successful UAV landings. This paper has designed a new control compensation method for an autonomous UAV landing platform to address the impact of complex sea conditions on the stability of UAV landing platforms. Firstly, the parallel Stewart platform was introduced as the landing platform, and its structure was analyzed with forward and inverse kinematic calculations conducted in Matlab to verify its accuracy. Secondly, a least-squares recursive AR prediction algorithm was designed to predict the future attitudes of ships under varying sea conditions. Finally, the prediction algorithm was combined with the platform's control strategy and a dual-sensor system was adopted to ensure the stability of the UAV landing process. The experimental results demonstrate that these innovative improvements enhanced the compensation accuracy by 59.6%, 60.3%, 48.4%, and 47.9% for the rolling angles of 5° and 10° and the pitching angles of 5° and 10°, respectively. Additionally, the compensation accuracy for the roll and pitch in sea states 2 and 5 improved by 51.2%, 59.4%, 58.7%, and 55.9%, respectively, providing technical support for UAV missions such as maritime rescue and exploration. [ABSTRACT FROM AUTHOR]

Abstract Recently, human motion prediction has gained significant attention and achieved notable success. However, current methods primarily rely on training and testing with ideal datasets, overlooking the impact of variations in the viewing distance and viewing angle, which are commonly encountered in practical scenarios. In this study, we address the issue of model invariance by ensuring robust performance despite variations in view distances and angles. To achieve this, we employed Riemannian geometry methods to constrain the learning process of neural networks, enabling the prediction of invariances using a simple network. Furthermore, this enhances the application of motion prediction in various scenarios. Our framework uses Riemannian geometry to encode motion into a novel motion space to achieve prediction with an invariant viewing distance and angle using a simple network. Specifically, the specified path transport square-root velocity function is proposed to aid in removing the view-angle equivalence class and encode motion sequences into a flattened space. Motion coding by the geometry method linearizes the optimization problem in a non-flattened space and effectively extracts motion information, allowing the proposed method to achieve competitive performance using a simple network. Experimental results on Human 3.6M and CMU MoCap demonstrate that the proposed framework has competitive performance and invariance to the viewing distance and viewing angle.

Abstract It is noted that Recurrent Neural Networks (RNNs), which are widely used in human prediction tasks, have achieved promising performance in motion prediction, owing to RNNs’ robust capacity for spatial-temporal sequence modeling. However, RNN-based methods suffer from error accumulation due to their step-by-step prediction mechanism. Therefore, in this paper, we propose a three-stage parallel prediction network, which guides the output generation of these three networks with different objectives. In particular, we leverage the high-dimensional information in these three networks to fuse new information to generate the final output. In addition, we also designed a fusion block based on GRU and attention mechanism to extract high-dimensional information more efficiently. Extensive experiments show that our approach outperforms most recent methods in both short and long-term motion predictions on Human 3.6M, CMU Mocap, and 3DPW.

The disturbance of the coal caving mechanism is the main method to solve the problem of coal falling into an arch during the top coal caving process, and the clearance in the caving mechanism is an important factor affecting its disturbance. To study the influence of clearance on the hydraulic supports, the dynamic analysis of hydraulic supports with two-level and three-level caving mechanisms is conducted considering the clearance. Based on Lagrange's equations of first kind, the mathematical model is constructed, and the equivalent spring stiffness of the oil cylinder is solved. The dynamic response of hydraulic support is studied by solving the mathematical model, and the relative position changes of the journal and bearing are compared. The research results prove that clearance has a better disturbance effect on the three-level caving mechanism. At the same time, the influence of clearance size and motion process on the motion error of the three-level caving mechanism is analyzed. Using the Kriging surrogate model and polynomial interpolation, the motion prediction model for the secondary tail beam with different clearance sizes is proposed, and the predicted values of the test samples are compared with the numerical solution values to prove the feasibility.

Video Assistant Referee (VAR) is used for line judging in tennis to improve the fairness of a game. However, currently, popular VAR systems are based on three-dimensional trajectories and do not take actual ball bounce marks into account, which may lead to misjudgment due to misalignment. Therefore, in this paper, we propose Apollon Mark, a bounce mark visualization system with preceding mirror control based on high-speed drop location prediction. Based on high-speed measurement of three-dimensional position and motion prediction, the proposed system enables a high-resolution camera to capture the ground texture both before and after the ball drops on the ground by directing the angle of view of the camera to the predicted drop location of the ball before it drops. The system then enables bounce mark visualization by simple differential processing between the before and after images with a template-matching image processing algorithm. We have quantitatively evaluated the validity of the proposed system through the simulation of drop location prediction for real-scale tennis ball trajectories and validated the bounce mark visualization algorithm under a dynamic angle of view using galvanometer mirrors.

The application of offshore wind power gangways ensures the normal operation of offshore wind turbines and provides reliable support for optimizing the boarding process, improving the operation and maintenance efficiency, and reducing the cost of wind power projects. This paper provides an overview of the development history and technological progress of gangways, and analyzes the main equipment in this field and its performance at home and abroad, as well as discussing the key technologies of offshore wind turbine gangways in such aspects as motion sensing and data analysis, compensation control and execution, structural design and reliability, and composite material application. It is pointed out that the parallel platform is the mainstream offshore gangway actuator; wave compensation is the core of equipment performance and the source of competitiveness; and advanced wave compensation technology depends on the development of sensor technology, artificial intelligence technology and other multidisciplinary development. This paper makes a comprehensive analysis for researchers and practitioners in the field of offshore wind power gangways, providing valuable references for subsequent technology exploration.

Multi-object tracking (MOT) aims at estimating locations and identities of objects in videos. Many modern multiple-object tracking systems follow the tracking-by-detection paradigm, consisting of a detector followed by a method for associating detections into tracks. Tracking by associating detections through motion-based similarity heuristics is the basic way. Motion models aim at utilizing motion information to estimate future locations, playing an important role in enhancing the performance of association. Recently, a large-scale dataset, DanceTrack, where objects have uniform appearance and diverse motion patterns, was proposed. With existing hand-crafted motion models, it is hard to achieve decent results on DanceTrack because of the lack of prior knowledge. In this work, we present a motion-based algorithm named TLtrack, which adopts a hybrid strategy to make motion estimates based on confidence scores. For high confidence score detections, TLtrack employs transformers to predict its locations. For low confidence score detections, a simple linear model that estimates locations through trajectory historical information is used. TLtrack can not only consider the historical information of the trajectory, but also analyze the latest movements. Our experimental results on the DanceTrack dataset show that our method achieves the best performance compared with other motion models.

Human motion prediction is a challenging task in human-centric computer vision, involving forecasting future poses based on historical sequences. Despite recent progress in modeling spatial-temporal relationships of motion sequences using complex structured graphs, few approaches have provided an adaptive and lightweight representation for varying graph structures of human motion. Taking inspiration from the advantages of MLP-Mixer, a lightweight architecture designed for learning complex interactions in multi-dimensional data, we explore its potential as a backbone for motion prediction. To this end, we propose a novel MLP-Mixer-based adaptive spatial-temporal pattern learning framework (M 2 AST). Our framework includes an adaptive spatial mixer to model the spatial relationships between joints, an adaptive temporal mixer to learn temporal smoothness, and a local dynamic mixer to capture fine-grained cross-dependencies between joints of adjacent poses. The final method achieves a compact representation of human motion dynamics by adaptively considering spatial-temporal dependencies from coarse to fine. Unlike the trivial spatial-temporal MLP-Mixer, our proposed approach can more effectively capture both local and global spatial-temporal relationships simultaneously. We extensively evaluated our proposed framework on three commonly used benchmarks (Human3.6M, AMASS, 3DPW MoCap), demonstrating comparable or better performance than existing state-of-the-art methods in both short and long-term predictions, despite having significantly fewer parameters. Overall, our proposed framework provides a novel and efficient solution for human motion prediction with adaptive graph learning. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Harbin Institute of Technology. Social Sciences Edition / Haerbin Gongye Daxue Xuebao. Shehui Kexue Ban is the property of Harbin Institute of Technology and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Robot joints are the main structure for controlling the motion of the machine body, where the motion state of them directly affects the performance of the industrial robot. Due to the difficulty of obtaining the joint torque information of industrial robots, it is very hard to monitor the motion state of them. Based on the velocity and force driven by current of motors, we propose a novel Bayesian optimization framework to predict the joint motion state of industrial robot in this paper. Based on the temporal correlation of joint current and the correlation between the current and motion state of joint, we use the LSTM and BiLSTM to regressing prediction of the current and state of joint motor first. Then, the Bayesian optimization method is used to adjust the hyperparameters of our network, which realize the analysis of the joint motor current under different motion states and improve the accuracy of the prediction of joint motion states. Finally, we design the joint current acquisition platform of industrial robot based on Hall current sensors, which can collect joint currents without contact and generate experimental dataset. Comparing with the popular intelligent methods, the results show that our Bayesian optimization framework realizes a more accurate prediction of motion state for the four-axis industrial robot on the basis of contact-less current acquisition. [ABSTRACT FROM AUTHOR]

This article focuses on motion prediction for point cloud sequences in the challenging case of deformable 3D objects, such as human body motion. First, we investigate the challenges caused by deformable shapes and complex motions present in this type of representation, with the ultimate goal of understanding the technical limitations of state-of-the-art models. From this understanding, we propose an improved architecture for point cloud prediction of deformable 3D objects. Specifically, to handle deformable shapes, we propose a graph-based approach that learns and exploits the spatial structure of point clouds to extract more representative features. Then, we propose a module able to combine the learned features in a adaptative manner according to the point cloud movements. The proposed adaptative module controls the composition of local and global motions for each point, enabling the network to model complex motions in deformable 3D objects more effectively. We tested the proposed method on the following datasets: MNIST moving digits, the Mixamo human bodies motions [15], JPEG [5] and CWIPC-SXR [32] real-world dynamic bodies. Simulation results demonstrate that our method outperforms the current baseline methods given its improved ability to model complex movements as well as preserve point cloud shape. Furthermore, we demonstrate the generalizability of the proposed framework for dynamic feature learning by testing the framework for action recognition on the MSRAction3D dataset [19] and achieving results on par with state-of-the-art methods. [ABSTRACT FROM AUTHOR]

Dynamic texts in scene videos provide valuable insights and semantic cues crucial for video applications. However, the movement of this text presents unique challenges, such as blur, shifts, and blockages. While efficient in tracking text, state‐of‐the‐art systems often need help when text becomes obscured or complicated scenes. This study introduces a novel method for detecting and tracking video text, specifically designed to predict the location of obscured or occluded text in subsequent frames using a tracking‐by‐detection paradigm. Our approach begins with a primary detector to identify text within individual frames, thus enhancing tracking accuracy. Using the Kalman filter, Munkres algorithm, and deep visual features, we establish connections between text instances across frames. Our technique works on the concept that when text goes missing in a frame due to obstructions, we use its previous speed and location to predict its next position. Experiments conducted on the ICDAR2013 Video and ICDAR2015 Video datasets confirm our method's efficacy, matching or surpassing established methods in performance. [ABSTRACT FROM AUTHOR]

Abstract Robot joints are the main structure for controlling the motion of the machine body, where the motion state of them directly affects the performance of the industrial robot. Due to the difficulty of obtaining the joint torque information of industrial robots, it is very hard to monitor the motion state of them. Based on the velocity and force driven by current of motors, we propose a novel Bayesian optimization framework to predict the joint motion state of industrial robot in this paper. Based on the temporal correlation of joint current and the correlation between the current and motion state of joint, we use the LSTM and BiLSTM to regressing prediction of the current and state of joint motor first. Then, the Bayesian optimization method is used to adjust the hyperparameters of our network, which realize the analysis of the joint motor current under different motion states and improve the accuracy of the prediction of joint motion states. Finally, we design the joint current acquisition platform of industrial robot based on Hall current sensors, which can collect joint currents without contact and generate experimental dataset. Comparing with the popular intelligent methods, the results show that our Bayesian optimization framework realizes a more accurate prediction of motion state for the four-axis industrial robot on the basis of contact-less current acquisition.

In this paper, we explore the application of data-driven predictive systems in enhancing unmanned aerial vehicle (UAV) control capabilities. We introduce a new model for predicting the motion of individual drones by utilizing fundamental flight control data. The model aims to improve the autonomy of individual drones and circumvent the complexity of traditional flight control systems, thus eliminating intricate nested controls. The proposed model lays the foundation for studying collective behaviours within a cluster of drones, thereby advancing the research into swarm behaviour exhibited by drones. The research findings demonstrate the potential of data-driven methods in the construction of UAV control systems. In particular, we here show a comparison of the prediction performances between two neural network architectures using real drone flight data involved in various kinds of motions. We explore the utility of using long short term memory (LSTM) and nonlinear autoregressive with exogenous inputs (NARX) family of nonlinear time series models in developing a virtual drone model using real experimental data.

Early detection of falls is important for reducing fall injuries. However, existing fall detection strategies mostly focus on reducing impact injuries rather than avoiding falls. This study proposed the concept of identifying "Imbalance Point" to warn the body imbalance, allowing sufficient time to recover balance. And if falling cannot be avoided, an impact sign is released by detecting the "Fall Point" prior to the impact. To achieve this goal, motion prediction model and balance recovery model are integrated into a spatiotemporal framework to analyze dynamic and kinematic features of body motion. Eight healthy young volunteers participated in three sets of experiment: Normal trial, Recovery trial and Fall trial. The body motion in the trials was recorded using Microsoft Azure Kinect. The results show that the developed algorithm for Fall Point detection achieved 100% sensitivity and 98.6% specificity, along with an average lead time of 297 ms. Moreover, Imbalance Point was successfully detected in all Fall trials, and the average time interval between Imbalance Point and Fall Point was 315 ms, longer than reported step reaction time for elderly (approximately 270 ms). The experiment results demonstrate that the developed algorithm have great potential for fall warning and protection in the elderly. [ABSTRACT FROM AUTHOR]

Reinforcement learning (RL)–based car-following (CF) control strategies have attracted significant attention in academia, emerging as a prominent research topic in recent years. Most of these control strategies focus solely on the motion status of the immediately preceding vehicle. However, with the development of vehicle-to-vehicle (V2V) communication technologies, intelligent vehicles such as connected autonomous vehicles (CAVs) can gather information about surrounding vehicles. Therefore, this study proposes an RL-based CF control strategy that takes multivehicle scenarios into account. First, the trajectories of two preceding vehicles and one following vehicle relative to the subject vehicle (SV) are extracted from a highD dataset to construct the environment. Then the twin-delayed deep deterministic policy gradient (TD3) algorithm is implemented as the control strategy for the agent. Furthermore, a sequence-to-sequence (seq2seq) module is developed to predict the uncertain motion statuses of surrounding vehicles. Once integrated into the RL framework, this module enables the agent to account for dynamic changes in the traffic environment, enhancing its robustness. Finally, the performance of the CF control strategy is validated both in the highD dataset and in two traffic perturbation scenarios. In the highD dataset, the TD3-based prediction CF control strategy outperforms standard RL algorithms in terms of convergence speed and rewards. Its performance also surpasses that of human drivers in safety, efficiency, comfort, and fuel consumption. In traffic perturbation scenarios, the performance of the proposed CF control strategy is compared with the model predictive controller (MPC). The results show that the TD3-based prediction CF control strategy effectively mitigates undesired traffic waves caused by the perturbations from the head vehicle. Simultaneously, it maintains the desired traffic state and consistently ensures a stable and efficient traffic flow. [ABSTRACT FROM AUTHOR]

Driving experience and anticipatory driving are essential skills for humans to operate vehicles in complex environments. In the context of autonomous vehicles, the software must offer the related features of scenario understanding and motion prediction. The latter feature of motion prediction is extensively researched with several competing large datasets, and established methods provide promising results. However, the incorporation of scenario understanding has been sparsely investigated. It comprises two aspects. First, by means of scenario understanding, individual assumptions of an object's behavior can be derived to adaptively predict its future motion. Second, scenario understanding enables the detection of challenging scenarios for autonomous vehicle software to prevent safety-critical situations. Therefore, we propose a method incorporating scenario understanding into the motion prediction task to improve adaptivity and avoid prediction failures. This is realized by an a priori evaluation of the scenario based on semantic information. The evaluation adaptively selects the most accurate prediction model but also recognizes if no model is capable of accurately predicting this scenario and high prediction errors are expected. The results on the comprehensive scenario library CommonRoad reveal a decrease in the Euclidean prediction error by 81.0 % and a 90.8 % reduction in mispredictions of our method compared to the benchmark model. [ABSTRACT FROM AUTHOR]

ObjectivesAiming at the challenge of accurate diving modeling of a smart float, an anti-saturation and noise least squares (ASNLS) algorithm is proposed in this paper to achieve diving multi-parameter identification and depth prediction. MethodsFirstly, the nonlinear motion characteristics of the smart float actuator were included in the gray box-based diving model to better fit the actual model, and the continuous diving motion equation was transformed into a discrete form to match the real-world discrete data sampling. Subsequently, the aforementioned discrete diving model was constructed into a correlation form to attenuate the influence of data noise. Finally, by adjusting the values of the covariance matrix, the designed diving parameter identification algorithm achieved resistance to data saturation. ResultsBased on the data of the South China Sea deep diving experiment of the smart float in 2021, diving model parameter identification and depth prediction are carried out. The results demonstrate that the ASNLS algorithm has faster convergence speed (31.8% higher than the least squares algorithm) and smaller depth prediction error (average absolute percentage errors less than 9% at different depths) than both the traditional least squares algorithm and supports the vector machine algorithm.ConclusionsConsequently, the ASNLS algorithm can provide effective support for the depth control and prediction of the smart float.

With the maturity of satellite imaging technology, satellite video has attracted more and more attention because of high spatial resolution and temporal resolution. Thus, object tracking as the main application task of satellite videos also becomes a popular research topic. Compared to natural videos, the difficulties of satellite videos object tracking mainly are caused by feature deficiency of small objectives, surroundings, occlusion (OCC), etc. In this article, based on a dual correlation filter (DCF) tracking framework, a new object tracking method for satellite videos is proposed to deal with the above-mentioned problems. For feature deficiency problem, the proposed super-resolution feature enhancement module could improve the feature discriminative ability of objects utilizing the prior information learned by a super-resolution network. For the OCC problem, a multilayer perceptron motion prediction network is designed to predict the position of objects when OCC occurs, which uses an online training strategy. Besides, KNN background subtraction also is introduced to reduce the interference of surroundings. Finally, these above-mentioned processes are combined together appropriately on the DCF tracking framework for better tracking results. Aiming to verify the performance of the proposed method, abundant experiments have been conducted on satellite video datasets. The experimental results show that the design of the proposed tracking method is effective and it also has a conspicuous advantage compared with some state-of-the-art methods.

Autonomous vehicles continue to advance, primarily due to the continuous development and improvement of deep learning methods. Motion prediction of road users is a function that humans perform naturally while driving, such as anticipating when a pedestrian is likely to cross an intersection or when a vehicle would like to merge into a lane. Similarly, the ability to predict the motion of other road users has become a key function for autonomous vehicles to optimally plan their path. This publication explores the recent advancements in motion prediction of road users propelled by both deep learning, as well as the release of large datasets. These datasets provide many challenging real-world scenarios which can be leveraged to train and test state-of-the-art deep learning models. This study provides an overview of motion prediction, state-of-the-art datasets, and state-of-the-art models along with a deep dive into their methods. Finally, there is a comparison of model performance, followed by recommendations into future research and concluding remarks.

To ensure the reliability of autonomous driving, the system must be capable of potential hazard identification and appropriate response to prevent accidents. This involves the prediction of possible developments in traffic situations and an evaluation of the potential danger of future scenarios. Precise Collision Risk Assessment (CRA) faces complex challenges due to uncertainties inherent in vehicle and road environmental conditions. This paper introduces a new CRA approach, the Multi-Dimensional Uncertainties-CRA (MDU-CRA), which integrates uncertainties related to driver behavior, sensor perception, motion prediction models, and road infrastructure into a comprehensive risk evaluation framework. The estimation of vehicle state is initiated using Extended Kalman Filtering (EKF) to capture uncertainties in sensor perception. Concurrently, a probabilistic motion prediction model based on Gaussian distributions has been developed, which considers the uncertainty in driver behavior. Subsequently, the uncertainty of the road structure is modeled using a truncated Gaussian distribution. Finally, collision risk is quantified as the future probability of collision through heuristic Monte Carlo (MC) sampling. This paper presents the results of two experiments Firstly, our proposed method is demonstrated to outperform the reference neural network-based method in terms of short-term motion prediction accuracy. Secondly, two driving scenarios are extracted and reconstructed from the Next Generation Simulation (NGSIM) dataset for validation and evaluation, i.e., an active lane-change scenario and an emergency braking scenario. In the domain of collision risk assessment, our approach consistently outperforms other evaluation methods. It exhibits the capability to perceive collision risks 2 to 5 seconds in advance, significantly reducing the probability of imminent collision incidents.

Dynamic multiscale graph neural networks have high motion prediction errors due to the low correlation between the internal joints of body parts and the limited perceptual fields. A multiscale global adaptive attention graph neural network for human motion prediction is proposed to reduce motion prediction errors. Firstly, a multi-distance partitioning strategy for dividing skeleton joint is proposed to improve the degree of temporal and spatial correlation of body joint information. Secondly, a global adaptive attention spatial temporal graph convolutional network is designed to dynamically enhance the network??s attention to the spatial temporal joints contributing to a motion in combination with global adaptive attention. Finally, this paper integrates the above two improvements into the graph convolutional neural network gate recurrent unit to enhance the state propagation performance of the decoding network and reduce prediction errors. Experimental results show that the prediction error of the proposed method is decreased on Human 3.6M dataset, CMU Mocap dataset and 3DPW dataset compared with state-of-the-art methods.

In this study, we introduce a novel method for an online trajectory replanning approach for fixed-wing Unmanned Aerial Vehicles (UAVs). Our method integrates moving obstacle predictions within a gradient-based optimization framework. The trajectory is represented by uniformly discretized waypoints, which serve as the optimization variables within the cost function. This cost function incorporates multiple objectives, including obstacle avoidance, kinematic and dynamic feasibility, similarity to the reference trajectory, and trajectory smoothness. To enhance prediction accuracy, we combine physics-based and pattern-based methods for predicting obstacle movements. These predicted movements are then integrated into the online trajectory replanning framework, significantly enhancing the system’s safety. Our approach provides a robust solution for navigating dynamic environments, ensuring both optimal and secure UAV operation.

In recent years, as intelligent ship-navigation technology has advanced, the challenge of accurately modeling and predicting the dynamic environment and motion status of ships has emerged as a prominent area of research. In response to the diverse time scales required for the prediction of ship motion, various methods for modeling ship navigation environments, ship motion, and ship traffic flow have been explored and analyzed. Additionally, these motion-prediction methods are applied for motion control, collision-avoidance planning, and route optimization. Key issues are summarized regarding ship-motion prediction, including online modeling of motion models, real ship validation, and consistency in modeling, optimization, and control. Future technology trends are predicted in mechanism-data fusion modeling, large-scale model, multi-objective motion prediction, etc. [ABSTRACT FROM AUTHOR]

Visual tracking and attribute estimation related to age or gender information of multiple person entities in a scene are mature research topics with the advent of deep learning techniques. However, when it comes to indoor images such as video sequences of retail consumers, data are not always adequate or accurate enough to essentially train effective models for consumer detection and tracking under various adverse factors. This in turn affects the quality of recognizing age or gender for those detected instances. In this work, we introduce two novel datasets: Consumers comprises 145 video sequences compliant to personal information regulations as far as facial images are concerned and BID is a set of cropped body images from each sequence that can be used for numerous computer vision tasks. We also propose an end-to-end framework which comprises CNNs as object detectors, LSTMs for motion forecasting of the tracklet association component in a sequence, along with a multi-attribute classification model for apparent demographic estimation of the detected outputs, aiming to capture useful metadata of consumer product preferences. Obtained results on tracking and age/gender prediction are promising with respect to reference systems while they indicate the proposed model's potential for practical consumer metadata extraction. [ABSTRACT FROM AUTHOR]

Over the past four decades, Arctic sea ice coverage has steadily declined. This loss of sea ice has amplified solar radiation and heat absorption from the ocean, exacerbating both polar ice loss and global warming. It has also accelerated changes in sea ice movement, posing safety risks for ship navigation. In recent years, numerical prediction models have dominated the field of sea ice movement prediction. However, these models often rely on extensive data sources, which can be limited in specific time periods or regions, reducing their applicability. This study introduces a novel approach for predicting Arctic sea ice motion within a 10-day window. We employ a Self-Attention ConvLSTM deep learning network based on single-source data, specifically optical flow derived from the Advanced Microwave Scanning Radiometer Earth Observing System 36.5 GHz data, covering the entire Arctic region. Upon verification, our method shows a reduction of 0.80 to 1.18 km in average mean absolute error over a 10-day period when compared to ConvLSTM, demonstrating its improved ability to capture the spatiotemporal correlation of sea ice motion vector fields and provide accurate predictions. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Frontiers of Computer Science & Technology is the property of Beijing Journal of Computer Engineering & Applications Journal Co Ltd. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Background and purpose: Active 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 methods: Seventy 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. Results: Using 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. Conclusion: It 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.

As an important interface between unmanned aerial vehicles (UAVs) and ships, the stability and motion control compensation technology of the shipborne UAV landing platform are paramount for successful UAV landings. This paper has designed a new control compensation method for an autonomous UAV landing platform to address the impact of complex sea conditions on the stability of UAV landing platforms. Firstly, the parallel Stewart platform was introduced as the landing platform, and its structure was analyzed with forward and inverse kinematic calculations conducted in Matlab to verify its accuracy. Secondly, a least-squares recursive AR prediction algorithm was designed to predict the future attitudes of ships under varying sea conditions. Finally, the prediction algorithm was combined with the platform’s control strategy and a dual-sensor system was adopted to ensure the stability of the UAV landing process. The experimental results demonstrate that these innovative improvements enhanced the compensation accuracy by 59.6%, 60.3%, 48.4%, and 47.9% for the rolling angles of 5° and 10° and the pitching angles of 5° and 10°, respectively. Additionally, the compensation accuracy for the roll and pitch in sea states 2 and 5 improved by 51.2%, 59.4%, 58.7%, and 55.9%, respectively, providing technical support for UAV missions such as maritime rescue and exploration.

With the development of computer vision and Unmanned Aerial Vehicles (UAVs) technology, visual object tracking has become an indispensable core technology for UAVs, and it has been widely used in both civil and military fields. Visual object tracking from the UAV perspective experiences interference from various complex conditions such as background clutter, occlusion, and being out of view, which can easily lead to tracking drift. Once tracking drift occurs, it will lead to almost complete failure of the subsequent tracking. Currently, few trackers have been designed to solve the tracking drift problem. Thus, this paper proposes a tracking algorithm based on motion prediction and block search to address the tracking drift problem caused by various complex conditions. Specifically, when the tracker experiences tracking drift, we first use a Kalman filter to predict the motion state of the target, and then use a block search module to relocate the target. In addition, to improve the tracker’s ability to adapt to changes in the target’s appearance and the environment, we propose a dynamic template updating network (DTUN) that allows the tracker to make appropriate template decisions based on various tracking conditions. We also introduce three tracking evaluation metrics: namely, average peak correlation energy, size change ratio, and tracking score. They serve as prior information for tracking status identification in the DTUN and the block prediction module. Extensive experiments and comparisons with many competitive algorithms on five aerial benchmarks, UAV20L, UAV123, UAVDT, DTB70, and VisDrone2018-SOT, demonstrate that our method achieves significant performance improvements. Especially in UAV20L long-term tracking, our method outperforms the baseline in terms of success rate and accuracy by 19.1% and 20.8%, respectively. This demonstrates the superior performance of our method in the task of long-term tracking from the UAV perspective, and we achieve a real-time speed of 43 FPS.

Background: Magnetic resonance imaging (MRI)‐guided radiotherapy with multileaf collimator (MLC)‐tracking is a promising technique for intra‐fractional motion management, achieving high dose conformality without prolonging treatment times. To improve beam‐target alignment, the geometric error due to system latency should be reduced by using temporal prediction. Purpose: To experimentally compare linear regression (LR) and long‐short‐term memory (LSTM) motion prediction models for MLC‐tracking on an MRI‐linac using multiple patient‐derived traces with different complexities. Methods: Experiments were performed on a prototype 1.0 T MRI‐linac capable of MLC‐tracking. A motion phantom was programmed to move a target in superior‐inferior (SI) direction according to eight lung cancer patient respiratory motion traces. Target centroid positions were localized from sagittal 2D cine MRIs acquired at 4 Hz using a template matching algorithm. The centroid positions were input to one of four motion prediction models. We used (1) a LSTM network which had been optimized in a previous study on patient data from another cohort (offline LSTM). We also used (2) the same LSTM model as a starting point for continuous re‐optimization of its weights during the experiment based on recent motion (offline+online LSTM). Furthermore, we implemented (3) a continuously updated LR model, which was solely based on recent motion (online LR). Finally, we used (4) the last available target centroid without any changes as a baseline (no‐predictor). The predictions of the models were used to shift the MLC aperture in real‐time. An electronic portal imaging device (EPID) was used to visualize the target and MLC aperture during the experiments. Based on the EPID frames, the root‐mean‐square error (RMSE) between the target and the MLC aperture positions was used to assess the performance of the different motion predictors. Each combination of motion trace and prediction model was repeated twice to test stability, for a total of 64 experiments. Results: The end‐to‐end latency of the system was measured to be (389 ± 15) ms and was successfully mitigated by both LR and LSTM models. The offline+online LSTM was found to outperform the other models for all investigated motion traces. It obtained a median RMSE over all traces of (2.8 ± 1.3) mm, compared to the (3.2 ± 1.9) mm of the offline LSTM, the (3.3 ± 1.4) mm of the online LR and the (4.4 ± 2.4) mm when using the no‐predictor. According to statistical tests, differences were significant (p‐value <0.05) among all models in a pair‐wise comparison, but for the offline LSTM and online LR pair. The offline+online LSTM was found to be more reproducible than the offline LSTM and the online LR with a maximum deviation in RMSE between two measurements of 10%. Conclusions: This study represents the first experimental comparison of different prediction models for MRI‐guided MLC‐tracking using several patient‐derived respiratory motion traces. We have shown that among the investigated models, continuously re‐optimized LSTM networks are the most promising to account for the end‐to‐end system latency in MRI‐guided radiotherapy with MLC‐tracking. [ABSTRACT FROM AUTHOR]