Distribution Of Trajectories Research Articles

Learning to search for parking like a human: A deep inverse reinforcement learning approach

The prevalence of on-street parking searches in urban downtown areas has led to significant externalities such as congestion, pollution, and collisions. Understanding the intricacies of parking search behavior is crucial for developing effective management strategies to mitigate these issues. Parking search is inherently a complex, sequential decision-making process, influenced by diverse driver preferences and dynamic urban environments. This study introduces a deep inverse reinforcement learning (DIRL) approach to model drivers’ parking search behavior. First, we constructed a high-fidelity parking simulation platform using Unity3D to replicate an urban road network, enabling the collection of 987 valid trajectories. We modeled the parking search process as a Markov decision process (MDP), with meticulously designed state-action pairs for accurate representation. Then, a maximum entropy-based DIRL model was developed to learn the reward function and search-for-parking policies of drivers. The experimental results demonstrate that the maximum entropy DIRL model significantly outperforms the traditional maximum entropy inverse reinforcement learning model, achieving a 19.0% improvement in accurately capturing final parking states and a 13.5% enhancement in characterizing overall trajectory distributions. Finally, we integrated these trained models into traditional traffic simulation systems to effectively observe the traffic state evolution with different parking search behaviors, providing valuable insights for optimizing urban traffic management strategies.

Modelling of Mass Transport in Fractured Crystalline Rock Using Velocity Interpolation and Cell-Jump Particle Tracking Methods

In this study, two particle tracking methods, velocity interpolation, and cell-jump, were employed to simulate tracer transport in fractured crystalline rock. The models, belonging to DECOVALEX-2023 Task F1, included one considering only the influence of deterministic (major) fractures, and another considering both deterministic and stochastic (background) fractures. The simulations involved converting fracture properties into equivalent hydraulic parameters for each three-dimensional grid, simulating steady-state flow fields, and evaluating transport parameters using particle tracking methods. Using transport parameters, one-dimensional transport pathways were simulated for evaluating mass transport of tracers considering non-reactive, decay, and adsorption. Moment analysis was then utilized to quantify breakthrough curves and compare the performance of the two particle tracking methods. The conclusion is that the cell-jump method, despite facing issues with numerical dispersion that results in a broader distribution of particle trajectories, demonstrates advantages in providing relative shorter mean breakthrough times and less temporal spreading compared to the velocity interpolation (VI) method in cases involving stochastic background fractures. Both methods are limited by the issue of particles entering the matrix due to the application of non-zero permeability for numerical convenience.

Inertial active Ornstein–Uhlenbeck particle in a non-linear velocity dependent friction

We explore the self-propulsion of an active Ornstein–Uhlenbeck particle with a non-linear velocity dependent friction. Using analytical approach and numerical simulation, we have exactly investigated the dynamical behaviour of the particle in terms of particle trajectory, position and velocity distribution functions in both underdamped as well as overdamped regimes of the dynamics. Analyzing the distribution functions, we observe that for a confined harmonic particle, with an increase in duration of self-propulsion, the inertial particle prefers to accumulate near the boundary of the confinement rather than the mean position, reflecting an activity induced bistability in the presence of nonlinear friction. On the other hand, in the overdamped or highly viscous regime, where the inertial influence is negligible, the sharp peak structure in the distribution across the mean position of the well reveals as usual trapping of the particle with increase in the persistent duration of activity. Moreover, for a free particle, using perturbation method, we have analytically computed the velocity distribution function in the vanishing limit of noise. The distribution interestingly shows the similar attributes as in case of a harmonic well, thus providing an additional effective confining mechanism that can be explained as a decreasing function of effective temperature. In this limit, the analytically computed distribution agrees well with the simulation results.

Molecular dynamics study on the enhancement of high‐temperature friction and mechanical properties of perfluoroelastomers by carbon nanomaterials

AbstractThe mechanical and frictional properties of pure perfluoroelastomer, graphene (GN)/perfluoroelastomer, and carbon nanotube (CNT)/perfluoroelastomer composite systems were investigated using molecular dynamics simulations. The influence of carbon nanomaterials on the glass transition temperature, mechanical properties, and frictional behavior of perfluoroelastomers at various temperatures was evaluated. The simulation results demonstrated that the addition of GN and CNTs increased the glass transition temperature of perfluoroelastomers to varying extents. CNTs enhanced the mechanical properties of perfluoroelastomers more effectively than GN, whereas GN provided superior improvement in frictional properties compared to CNTs. Analysis of stress–strain distribution, atomic motion trajectory, interaction energy, mass density, and temperature distribution, bond orientation parameters, radial distribution function (RDF), mean square displacement (MSD), and diffusion coefficients revealed the mechanisms and differences by which GN and CNTs enhance the mechanical and frictional properties of perfluoroelastomers.Highlights Carbon nanomaterials enhance perfluoroelastomer's high‐temperature performance. CNTs provide higher mechanical strength but lower wear resistance. GN nanosheets outperform CNTs in reducing friction and wear. Molecular dynamics simulations reveal distinct enhancement mechanisms.

Vehicle Trajectory Prediction Using Residual Diffusion Model Based on Image Information

In automatic driving, accurate prediction of vehicle trajectory is the key to achieve automatic driving, and multi-vehicle joint trajectory prediction has become an important part of modern human-computer interaction systems such as automatic driving. In order to better predict vehicle trajectories, we propose a new residual diffusion model to infer the joint distribution of future multi-vehicle trajectories. This approach has several major advantages. First, the model is able to learn multiple probability distributions from trajectory data to obtain potential outcomes for vehicles to multiple future trajectories. Secondly, in order to integrate the motion characteristics of multiple vehicles in the same scene, we use the method of combining the reference denoising and multiple residual denoising to improve the model performance and prediction speed. Finally, on this basis, a general trajectory constraint function is introduced, so that the generated trajectories of multiple vehicles will not collide with each other. We perform a rich experimental comparison of various existing methods on the NGSIM dataset and demonstrate that the proposed algorithm achieves a 26% improvement on mAP.

A general image classification model for agricultural machinery trajectory mode recognition

Field-road trajectory classification is a crucial task for agricultural machinery behavior mode recognition, aiming to distinguish field operation mode and road driving mode automatically. However, the imbalanced distribution of agricultural machine trajectories brings challenges for the field-road trajectory classification task. Additionally, most existing field-road trajectory classification methods have certain shortcomings. For instance, they encounter difficulties in accurately representing the state of agricultural machinery movement using the current features. The data transformation process often leads to information loss, and the model’s generalization capabilities are limited. The performance of the models is constrained by each of these elements. To address these shortcomings, this paper introduces a general image classification model for agricultural machinery trajectory mode recognition named ATRNet. First, to address the issue of imbalanced field-road proportions in agricultural machinery trajectory data, a Conditional Tabular Generative Adversarial Network (CTGAN) is employed to generate quasi trajectories, balancing the distribution of positive and negative samples in the data. This step aims to eliminate biases during the model training process. Second, to accurately characterize the motion status of agricultural machinery, we propose a multiangle feature enhancement method to extract rich spatiotemporal features from trajectory data. Finally, different from conventional field-road trajectory classification models that primarily rely on spatial and temporal information for identifying trajectories, we present a lossless trajectory data representation paradigm. This paradigm maps each trajectory point into a “feature map” and uses an image classification model to capture latent feature representations of trajectory points for the recognition of different behavior modes of agricultural machinery. This paradigm can generalize image classification networks to the field-road trajectory classification task, providing a general vision model solution for agricultural machinery trajectory mode recognition. To validate the effectiveness of the ATRNet model, experiments were conducted on real corn and wheat harvester trajectory datasets. The results demonstrate that the proposed model achieves remarkable performance improvements over the state-of-the-art (SOTA) models. In the corn harvester trajectory dataset, ATRNet achieves an accuracy of 92.36% and an F1-score of 92.34%, surpassing existing SOTA models by 3.12% and 12.46%, respectively. Similarly, in the wheat harvester trajectory dataset, ATRNet achieves an accuracy of 92.36% and an F1-score of 92.33%, outperforming the existing optimal algorithm by 4.76% and 18.18%, respectively.

Analysis of microplastic behaviors in river-type lakes using a quasi-three-dimensional microplastic transport model

Paldang Lake in South Korea is a river-type lake characterized by weak temperature stratification and a short water residence time. It exhibits complex flow structures resulting from the convergence of two major rivers, the North Han River (NHR) and the South Han River (SHR), and one small river, the Gyeongan Stream. Analyzing the spatio-temporal variations in microplastic (MP) concentration in Paldang Lake is important because of its significance as a water supply source for the metropolitan area. In this study, a quasi-three-dimensional MP transport model (MPT-Q3D) based on the Lagrangian particle-tracking technique was developed to analyze the trajectory and concentration distribution of MPs driven by shear dispersion in river-type lakes. Using the simulation results of MPT-Q3D, which were validated with MP monitoring data and EFDC simulations, we investigated the spatio-temporal variations in MP concentration during both low-flow and flood-flow periods. Simulation results showed that horizontal transport by shear flow contributed more to MP transport than vertical transport, including settling movements based on MP density. Thus, the MP transport during the low-flow period was significantly influenced by the recirculating flow induced by hydropeaking of the NHR, as well as flow blockage caused by the strong inflow momentum from the SHR. During the flood-flow period, flood discharge from the three tributary rivers had an equal influence on MP transport and the variation in MP concentrations around the water intake facilities. Furthermore, the MP residence time was primarily affected by variations in the flow characteristics of the inflowing tributaries rather than the properties of the MP types. Analysis of the influx and efflux of MPs in Paldang Lake revealed that polypropylene was the predominant constituent, accounting for 53.9 % and 75.2 % of all residual MPs during the low-flow and flood-flow periods, respectively.

Advancing Vehicle Trajectory Prediction: A Probabilistic Approach Using Combined Sequential Models

Abstract This paper addresses the critical need to quantify vehicle trajectory uncertainty in autonomous driving under environmental variability. We focus on predicting the posterior distribution of vehicle trajectories over a fixed horizon, given an initial state and a sequence of actions. We propose and compare three approaches: a probabilistic seq2seq model based on stochastic variational Gaussian processes, sequential Monte Carlo simulation with a single-step Gaussian process model, and a hybrid model that leverages the strengths of both methods. Each approach incorporates a baseline vehicle kinematics model to enhance stability and convergence. We evaluate these methods using a dataset generated from the CARLA simulator, assessing both point error metrics and probabilistic prediction metrics. This research introduces novel approaches to quantifying vehicle trajectory uncertainty through various uncertainty quantification techniques, with the goal of improving the safety and reliability of autonomous vehicle control systems.

Characteristics analysis and safety evaluation of vehicles turning over long distances

In order to explore the motion and safety characteristics of the U-turn vehicles at long-range U-turn intersections, based on the measured data of the U-turn vehicles at typical locations, we analyzed the characteristics of the U-turn vehicle's trajectories and traffic parameters with the help of probability theory and statistical knowledge, and established the U-turn trajectory model. We used Traffic Conflict Technology (TCT) and take Post Encroachment Time (PET) as the indicator to study the feature of vehicles security of - turn vehicles. Three conclusions are drawn from this study. First, there is a significant difference in the headway between long-range -turn vehicles and those at intersections without opening: the stability of the headway of long-range -turn vehicles is weak, and the release of vehicles is seriously disturbed by the turning flow; Secondly, when turning around at a long interval u-turn intersection, the starting position of vehicle turning is dispersed and the distribution of U-turn trajectories is quadratic distribution. Third, the setting of long interval turning lanes will increase the interference between traffic flows at intersections, thus increasing the number of conflicts with vehicles in the opposite direction. Cross-conflict accounts for the majority of conflicts in long interval U-turn intersections, and rear-end collision is the main conflict type in U-turn intersections without opening. The above research conclusions are helpful to improve the safety of long interval U-turn intersections.

New Evidence of Healthier Aging: Positive Cohort Effects on Verbal Fluency.

Cross-sectional studies have shown improvements in cognition in later-born cohorts. However, it remains unclear whether these cohort effects extend beyond cognitive levels and are also detectable in the rate of age-related cognitive decline. Additionally, evidence is scarce on the presence and consistency of cohort effects throughout different segments of the distribution of cognitive trajectories. This study evaluates the existence and variability of cohort effects across the entire distribution of aging-related trajectories of verbal fluency. With this purpose, we develop sex and education-adjusted longitudinal norms of verbal fluency using data from 9 waves of the English Longitudinal Study of Aging (ELSA) by fitting quantile mixed models. The effect of age was modeled using splines to assess birth cohort effects, after grouping individuals in 5-year groups from 1920 to 1950 according to their age at study entry. To test for possible cohort effects across the 10th, 50th, and 90th quantiles, the coefficients associated with the splines were allowed to vary among cohorts. Our results suggest that, consistently across longitudinal quantiles, decline in verbal fluency across age is less pronounced for later-born individuals (p < .001), supporting the hypothesis of cohort effects. Additionally, we also found that quantiles of verbal fluency at any age are shifted upwards in later-born cohorts compared to those in earlier-born cohorts. These results enhance our understanding of cognitive decline in older adults by demonstrating that cohort effects on cognition are observable both cross-sectionally and longitudinally, affecting the entire range of verbal fluency trajectories.

Field evidence and indicators of rockfall fragmentation and implications for mobility

Rockfall fragmentation can play an important role in hazard studies and the design of protective measures. However, the current lack of modeling tools that incorporate rock fragmentation mechanics is a limitation to enhancing studies and design. This research investigates the fragmentation patterns of rockfalls and analyzes the resulting distribution of fragment sizes within corresponding rockfall deposits. We focus on small rock fragments, which provide insights into the dynamics of the rockfall event and can be used as input for numerical modeling. We analyzed multiple rockfall events from locations worldwide, each exhibiting different degrees of fragmentation. Using image analysis techniques, we mapped all visible blocks, determined their volumes, and measured the distances they travelled from the initial point of impact. A key finding is the identification of three indicators of fragmentation. First, in cases where fragmentation was largely absent, we observed a trend of increasing block size with distance from the impact point or source area, which aligns with previously published findings. However, for energetic rockfall events characterized by intense fragmentation, we observed that small fragments exhibited longer travel distances compared to larger fragments. This distinction allowed us to differentiate blocks primarily resulting from the disaggregation process from those primarily resulting from dynamic fragmentation, with implications for rockfall mobility. Second, although the size distribution of rockfall deposits exhibits a power-law scaling for volumes larger than a minimum size threshold corresponding to a rollover of the distribution, in some case studies a deviation from power-law scaling is observed, indicating a process of larger block comminution due to fragmentation. Third, we found that rockfalls with fragmentation experience reduced mobility, indicated by higher reach angles, and higher lateral dispersion showing a wider distribution of trajectories. We interpret these findings as being directly related to the energy-consuming nature of fragmentation, which prevents farther deposition of fragmented rock blocks.

Effect of oscillation parameters on weld surface morphology during laser beam oscillation welding of stainless steel

During laser beam oscillation welding (LBOW), the weld surface morphology is important because it influences the appearance of the workpiece and reflects the welding quality, and the oscillation parameters have an important effect on the weld surface morphology. Therefore, a combination of experimental research and numerical simulation is conducted to investigate the influence of oscillation parameters, including oscillation mode, oscillation frequency, and oscillation amplitude, on the weld surface morphology during LBOW. The calculation encompasses the oscillation trajectory, combined velocity, and energy distribution on the weld surface, varying with different oscillation parameters. A three-dimensional heat transfer and fluid flow model with the oscillation heat source is established to simulate the temperature and fluid flow behavior inside the weld pool. When the heat input is constant, an increase in oscillation frequency results in a decrease in weld width and a smoother surface. As the oscillation frequency increases, the weld width decreases, the surface becomes smooth, and the energy distribution on the weld surface becomes more uniform. The fact that the period of the surface pattern is almost equal to the oscillation period indicates a direct correlation between the formation of weld surface pattern and the oscillation trajectory in LBOW. Among all the oscillation modes, the minimum combined velocity which is equal to the welding velocity occurs at the inflection point in the linear oscillation mode, and the variation in combined velocity of the laser beam is the smallest for circular oscillation mode. The surface pattern is primarily influenced by the oscillation trajectory and fluid flow within the weld pool. Under conditions of lower oscillation frequencies or larger oscillation amplitudes, the fluid flow is weaker, and the surface pattern is primarily influenced by the oscillation trajectory. Conversely, when the fluid flow is stronger, the surface pattern is mainly determined by the fluid flow within the weld pool under conditions of higher oscillation frequencies or smaller oscillation amplitudes. This study explains the formation mechanisms of weld surface patterns and provides insights and guidance for selecting optimal oscillation parameters in LBOW.

Complex dynamics in nonlinear small time-delayed optoelectronic oscillator and application in fast reservoir computing and pulse generation

We investigated a time-delayed optoelectronic oscillator (OEO) that displays a wide range of complex dynamic behavior under small time delay. The phase-space trajectory distributions in different dynamic regimes were compared which brings a new perspective on the underlying mechanism of the transition process. It was found that bifurcation is always possible no matter how small the time delay is even if the universal adiabatic approximation model is invalid. Hereby we proposed a versatile simple oscillator which has a potential capacity as memory carrier and high-dimensional state spatial mapping ability that brings 1000 times computing-efficiency improvements of reservoir computing over the large time delay one. Furthermore, we demonstrated a new approach for a tunable optoelectronic pulse generator (repetition rate at 0.2 MHz and 0.25 GHz) which depends critically on time-delayed input electrical pulse. The proposed oscillator is also a promising system for the applications of fast chaos-based communication.

Competition among recombination pathways in single FAPbBr3 nanocrystals.

Single particle level microscopy of immobilized FAPbBr3 nanocrystals (NCs) has elucidated the involvement of different processes in their photoluminescence (PL) intermittency. Four different blinking patterns are observed in the data from more than 100 NCs. The dependence of PL decays on PL intensities brought out in fluorescence lifetime intensity distribution (FLID) plots is rationalized by the interplay of exciton- and trion-mediated recombinations along with hot carrier (HC) trapping. The high intensity-long lifetime component is attributed to neutral exciton recombination, the low intensity-short lifetime component is attributed to trion assisted recombination, and the low intensity-long lifetime component is attributed to hot carrier recombination. Change-point analysis (CPA) of the PL blinking data reveals the involvement of multiple intermediate states. Truncated power law distribution is found to be more appropriate than power law and lognormal distribution for on and off events. Probability distributions of PL trajectories of single NCs are obtained for two different excitation fluences and wavelengths (λex = 400, 440nm). Trapping rate (kT) prevails at higher power densities for both excitation wavelengths. From a careful analysis of the FLID and probability distributions, it is concluded that there is competition between the HC and trion assisted blinking pathways and that the contribution of these mechanisms varies with excitation wavelength as well as fluence.

Influence of sea ice on ship routes and speed along the Arctic Northeast Passage

The accelerated melting of sea ice and the growing influx of ships in the Arctic pose significant challenges to navigational safety and the resilient development of the Arctic routes. This study aims to explore the influence of sea ice on ship routes and speed by analyzing the spatiotemporal correlation of Sea Ice Concentration (SIC), Sea Ice Thickness (SIT), Sea Ice Volume (SIV), and Automatic Identification System (AIS) data along the Northeast Passage (NEP) in 2022. Our findings indicate that during the navigation window period (July–October), 115 ships traversed the NEP. It was observed that ships preferred navigating at speeds ranging from 6kn to 14kn when the SIC was below 10% and the SIT was less than 0.1m. Compared with cargo ships and tankers, the trajectory and speed distributions of other ships show greater differences under the influence of sea ice. Furthermore, our results suggest that the SIV, compared to the SIC and SIT, could better reflect the influence of sea ice on ship navigation. These insights are crucial for developing management strategies in Arctic routes.

Spatio-Temporal-Interaction Graph Neural Networks for Multi-Agent Trajectory Prediction

Abstract For intelligent transportation systems, accurately forecasting the future trajectories of multiple agents is pivotal. Considering the increased diversity of agents within a scene, in order to capture and model the variations in their appearance, motion status, behavioral patterns, and interrelationships, we propose a simple yet effective framework based on Spatio-Temporal-Interaction Graph Neural Networks. Specifically, a Multi-Class Agent Encoder is meticulously tailored to the specific class of each agent to distill pertinent information from their motion attributes and historical trajectories. Subsequently, a Spatio-Temporal-Interaction Graph Attention Module is constructed to productively represent and learn the complex, dynamic interactions. Finally, a Multimodal Trajectory Generation Module is customized, and a learnable diversity sampling function is introduced to map the features of each agent to a set of potential variables, so as to capture the multimodal distribution of future trajectories. Empirical evaluations on the ETH/UCY and KITTI datasets reveal that our method can efficiently improve the accuracy of trajectory prediction.

Modulation of H atom photoelectron momentum distribution by chirped laser pulses of different wavelengths

The wavelength dependence of the holographic interference structure in the photoelectron momentum distribution (PMD) of hydrogen atoms under a few-cycle chirped laser field is investigated using a two-dimensional (2D) semiclassical two-step (SCTS) model. Under the constraint of a fixed chirp parameter, we observe an amplification in the impact of the chirp parameter on the laser field waveform as the wavelength increases. This enhancement facilitates effective modulation of the ionization time window. As the wavelength reaches a certain threshold, the double-slit interference pattern undergoes an observable expansion towards the left. Furthermore, based on the variations in X-axis momentum (Px) concerning ionization time at different wavelengths and the corresponding distribution of classical electron trajectories at those wavelengths, we present a more comprehensive explanation for the expansion of interference patterns. This expansion can be attributed to changes in wavelength, which result in modifications to the laser field’s waveform, consequently leading to a significant acceleration of ionized electrons when exposed to chirped laser pulses. It is worth noting that by tuning the negative chirp parameter, the leftward expansion of the double-slit interference pattern can also be achieved, particularly when shorter wavelengths of chirped laser pulses are employed.

Laminated insulated glass units under blast loads: Experimental and numerical study

Laminated insulating glass units (IGUs), commonly used in modern buildings, are prone to shattering from explosive loads, causing significant damage and loss. Existing research on IGUs is constrained by a lack of experimental validation and comprehensive numerical analyses, as large-scale explosion tests are rarely reported due to safety and cost concerns. Furthermore, previous modeling efforts often oversimplify critical factors such as the supporting structure, component interactions, strain rate effects, hourglass energy, and failure criteria, leading to an inadequate representation of glass behavior under explosive loads. This paper addresses these gaps by conducting large-scale explosive test on glass system and utilizing sophisticated nonlinear dynamic analysis with LS-DYNA for simulation. The research considers the influence of the support system and incorporates detailed considerations of component interactions, constitutive relationships, and strain rate effects in the numerical simulations. The results confirm that the proposed model successfully replicates the crack patterns and damage observed in experimental test. Also, it accurately predicts the trajectory, quantity, and size distribution of glass fragments. The proposed modeling approach offers a more informed assessment of damage and improves the evaluation of glass safety in practical applications.

A Novel Learning-Based Trajectory Generation Strategy for a Quadrotor.

In this article, a learning-based trajectory generation framework is proposed for quadrotors, which guarantees real-time, efficient, and practice-reliable navigation by online making human-like decisions via reinforcement learning (RL) and imitation learning (IL). Specifically, inspired by human driving behavior and the perception range of sensors, a real-time local planner is designed by combining learning and optimization techniques, where the smooth and flexible trajectories are online planned efficiently in the observable area. In particular, the key problems in the framework, temporal optimality (time allocation), and spatial optimality (trajectory distribution) are solved by designing an RL policy, which provides human-like commands in real-time (e.g., slower or faster) to achieve better navigation, instead of generating traditional low-level motions. In this manner, real-time trajectories are calculated using convex optimization according to the efficient and accurate decisions of the RL policy. In addition, to improve generalization performance and to accelerate the training, an expert policy and IL are employed in the framework. Compared with existing works, the kernel contribution is to design a real-time practice-oriented intelligent trajectory generation framework for quadrotors, where human-like decision-making and model-based optimization are integrated to plan high-quality trajectories. The results of comparative experiments in known and unknown environments illustrate the superior performance of the proposed trajectory generation strategy in terms of efficiency, smoothness, and flexibility.

Transverse vibration analysis and active disturbance rejection decoupling control of rain erosion whirling arm

The transverse vibration caused by the mass eccentricity of the whirling arm and the unbalanced magnetic pull of the motor rotor is a critical factor that affects the stability and safety of the rain erosion whirling arm. To investigate the transverse vibration characteristics of the rain erosion whirling arm during operation, a four-degree-of-freedom transverse vibration model of the rotor-rain erosion whirling arm is established using the lumped mass method. The differential equation of motion of the rotor-rain erosion whirling arm, considering the influence of mass eccentricity and unbalanced magnetic pull, is then established based on the Lagrange equation. This equation is numerically solved using the Runge-Kutta algorithm to investigate the axis trajectory and amplitude distribution of the rotor-rain erosion whirling arm. To reduce the fatigue damage caused by vibration, the active disturbance rejection decoupling control method is employed. Furthermore, the parameters of the extended state observer are adjusted using pole assignment and bandwidth. The simulation results reveal that the vibration amplitude of the rotating blade can be reduced to within 0.2 mm, referring to the BSS7393 test standard. Additionally, experimental research is conducted to compare the vibration of the rain erosion whirling arm before and after implementing the active disturbance rejection decoupling control. The results demonstrate that the active disturbance rejection decoupling control method effectively suppresses the vibration of the rain erosion whirling arm, illustrating its feasibility and effectiveness for system vibration control.