Search Results (1,822)

With increasing incidents due to congestion at events, effective pedestrian guidance has become a critical safety concern. Recent research has explored the application of reinforcement learning to crowd simulation, where agents learn optimal actions through trial and error to maximize rewards based on environmental states. This study investigates the use of reinforcement learning and simulation techniques to mitigate pedestrian congestion through improved guidance systems. We employ the Multi-Agent Deep Deterministic Policy Gradient (MA-DDPG), a multi-agent reinforcement learning approach, and propose an enhanced method for learning the Q-function for actors within the MA-DDPG framework. Using the Mojiko Fireworks Festival dataset as a case study, we evaluated the effectiveness of our proposed method by comparing congestion levels with existing approaches. The results demonstrate that our method successfully reduces congestion, with agents exhibiting superior cooperation in managing crowd flow. This improvement in agent coordination suggests the potential for practical applications in real-world crowd management scenarios. Full article

We proposed a novel approach to investigate how problem-solving strategies, identified using response time and eye-tracking data, can impact individuals’ performance on the Object Assembly (OA) task. To conduct an integrated assessment of spatial reasoning ability and problem-solving strategy, we applied the Multimodal Joint-Hierarchical Cognitive Diagnosis Model (MJ-DINA) to analyze the performance of young students (aged 6 to 14) on 17 OA items. The MJ-DINA model consists of three sub-models: a Deterministic Inputs, Noisy “and” Gate (DINA) model for estimating spatial ability, a lognormal RT model for response time, and a Bayesian Negative Binomial (BNF) model for fixation counts. In the DINA model, we estimated five spatial cognitive attributes aligned with problem-solving processes: encoding, falsification, mental rotation, mental displacement, and intractability recognition. Our model fits the data adequately, with Gelman–Rubin convergence statistics near 1.00 and posterior predictive p-values between 0.05 and 0.95 for the DINA, Log RT, and BNF sub-models, indicating reliable parameter estimation. Our findings indicate that individuals with faster processing speeds and fewer fixation counts, which we label Reflective-Scanner, outperformed the other three identified problem-solving strategy groups. Specifically, sufficient eye movement was a key factor contributing to better performance on spatial reasoning tasks. Additionally, the most effective method for improving individuals’ spatial task performance was training them to master the falsification attribute. This research offers valuable implications for developing tailored teaching methods to improve individuals’ spatial ability, depending on various problem-solving strategies. Full article

Dynamic path planning enables vehicles to autonomously navigate in unknown or continuously changing environments, thereby reducing reliance on fixed maps. Deep reinforcement learning (DRL), with its superior performance in handling high-dimensional state spaces and complex dynamic environments, has been widely applied to dynamic path planning. Traditional DRL methods are prone to capturing unnecessary noise information and irrelevant features during the training process, leading to instability and decreased adaptability of models in complex dynamic environments. To address this challenge, we propose a dynamic path-planning method based on our Causal State-Masking Twin-delayed Deep Deterministic Policy Gradient (CSM-TD3) algorithm. CSM-TD3 integrates a causal inference mechanism by introducing dynamic state masks and intervention mechanisms, allowing the policy network to focus on genuine causal features for decision optimization and thereby enhancing the convergence speed and generalization capabilities of the agent. Furthermore, causal state-masking DRL allows the system to learn the optimal mask configurations through backpropagation, enabling the model to adaptively adjust the causal features of interest. Extensive experimental results demonstrate that this method significantly enhances the convergence of the TD3 algorithm and effectively improves its performance in dynamic path planning. Full article

Contemporary fluid motion modelling techniques, including the phenomenon of liquid sloshing in tanks, are increasingly associated with the use of artificial intelligence methods. In addition to the still frequently used traditional analysis methods and techniques, such as FEM, CFD, VOF and FSI, there is an increasing number of publications that use elements of artificial intelligence. Among others, artificial neural networks and deep learning techniques are used here in the field of prediction and approximation, as well as genetic and other multi-agent algorithms for optimization. This article analyses of the current state of research using the above techniques and the possibilities and main directions of their further development. Full article

This paper presents the application of reinforcement learning algorithms in the tuning of PID controllers for the control of some classes of continuous nonlinear systems. Tuning the parameters of the PID controllers is performed with the help of the Twin Delayed Deep Deterministic Policy Gradient (TD3) algorithm, which presents a series of advantages compared to other similar methods from machine learning dedicated to continuous state and action spaces. The TD3 algorithm is an off-policy actor–critic-based method and is used as it does not require a system model. Double Q-learning, delayed policy updates and target policy smoothing make TD3 robust against overestimation, increase its stability, and improve its exploration. These enhancements make TD3 one of the state-of-the-art algorithms for continuous control tasks. The presented technique is applied for the control of a biotechnological system that has strongly nonlinear dynamics. The proposed tuning method is compared to the classical tuning methods of PID controllers. The performance of the tuning method based on the TD3 algorithm is demonstrated through a simulation, illustrating the effectiveness of the proposed methodology. Full article

The excessive vibration or collapse of a transmission tower-line (TTL) system under seismic excitation can result in significant losses. Viscoelastic material dampers (VMDs) have been recognized as an effective method for structural vibration mitigation. Most existing studies have focused solely on the dynamic analysis of TTL systems with control devices under deterministic seismic excitations. Studies focusing on the nonstationary stochastic control of TTL systems with VMDs have not been reported. To this end, this study proposes a comprehensive analytical framework for the nonstationary stochastic responses of TTL systems with VMDs under stochastic seismic excitations. The analytical model of the TTL system is formulated using the Lagrange equation. The six-parameter model of VMDs and the vibration control method are established. Following this, the pseudo-excitation method (PEM) is applied to compute the stochastic response of the controlled TTL system under nonstationary seismic excitations, and a probabilistic framework for analyzing extreme value responses is developed. A real TTL system in China is selected to verify the validity of the proposed method. The accuracy of the proposed framework is validated based on the Monte Carlo method (MCM). A detailed parametric investigation is conducted to determine the optimal damper installation scheme and examine the effects of the service temperature and site type on stochastic seismic responses. The nonstationary stochastic seismic responses of the TTL system are consistent with those based on MCM, validating the accuracy of the proposed analytical framework. VMDs can effectively suppress the structural dynamic responses, with particularly stable control over displacement. The temperature and site type have a notable influence on the stochastic seismic responses of the TTL system. The research findings provide important references for improving the seismic performance of VMDs in TTL systems. Full article

The cooperative task assignment problem with time windows for heterogeneous multiple unmanned aerial vehicles is an attractive complex combinatorial optimization problem. In reality, unmanned aerial vehicles’ fuel consumption exhibits uncertainty due to environmental factors or operational maneuvers, and accurately determining the probability distributions for these uncertainties remains challenging. This paper investigates the heterogeneous multiple unmanned aerial vehicle cooperative task assignment model that incorporates time window constraints under uncertain environments. To model the time window constraints, we employ the big-M method. To address the uncertainty in fuel consumption, we apply an adjustable robust optimization approach combined with duality theory, which allows us to derive the robust equivalent form and transform the model into a deterministic mixed-integer linear programming problem. We conduct a series of numerical experiments to compare the optimization results across different objectives, including maximizing task profit, minimizing total distance, minimizing makespan, and incorporating three different time window constraints. The numerical results demonstrate that the robust optimization-based heterogeneous multiple unmanned aerial vehicle cooperative task assignment model effectively mitigates the impact of parameter uncertainty, while achieving a balanced trade-off between robustness and the optimality of task assignment objectives. Full article

With the advent of large-scale electronic transportation, the construction of electric vehicle charging stations (EVCSs) has increased. The stochastic characteristic of the charging power of EVCSs leads to a risk of destabilization of the DC distribution network when there is a high degree of power electronification. Current deterministic stability analysis methods are too complicated to allow for brief descriptions of the effect of probabilistic characteristics of EVCSs on stability. This paper develops a probabilistic small-signal stability analysis method. Firstly, the probabilistic information of the system is obtained by combining the s-domain nodal impedance matrix based on the point estimation method. Then, the probability function of stability is fitted using the Cornish–Fisher expansion method. Finally, a comparison experiment using Monte Carlo simulation demonstrates that this method performs well in balancing accuracy and computational efficiency. The effects of line parameters and system control parameters on stability are investigated in the framework of probabilistic stability. This will provide a probabilistic perspective on the design of more complex power systems in the future. Full article

Nuclear power plant risk assessments rely on conservative deterministic criteria for core-damage determination despite significant advancements in plant response and system analyses. This study proposes a probabilistic approach to determine fuel rod fracture during loss-of-coolant accidents (LOCAs) in light-water reactors, addressing the need for more rational and realistic assessments. The methodology integrates a fuel rod fracture probability estimation model with best-estimate-plus-uncertainty analysis of plant response, utilizing the stress–strength model and Monte Carlo simulations. Both stress and strength distributions are estimated through Bayesian statistical modeling, with numerical integration techniques implemented to enhance accuracy for low-frequency events. The application of this approach to a virtual dataset demonstrated that while conventional deterministic methods indicated definitive rod fracture, our probabilistic analysis revealed a more realistic fracture probability of 15.1%. This significant finding highlights the potential reduction in assessment conservatism. The proposed methodology enables a transition from conservative binary evaluations to more realistic probabilistic assessments of core damage, providing more accurate risk insights for decision-making. Full article

This article presents a method for solving the inverse problem of computed tomography using an incomplete dataset. The problem focuses on reconstructing spatial objects based on the data collected from transmitters and receivers (referred to as projection vectors). The novelty of the proposed approach lies in combining two types of algorithms, namely heuristic and deterministic. Specifically, Artificial Bee Colony (ABC) and Jellyfish Search (JS) algorithms were utilized and compared as heuristic methods, while the deterministic methods were based on the Hooke–Jeeves (HJ) and Nelder–Mead (NM) approaches. By merging these techniques, a hybrid algorithm was developed, integrating the strengths of both heuristic and deterministic algorithms. The proposed hybrid algorithm proved to be approximately five to six times faster than an approach relying solely on metaheuristics while also providing more accurate results. In the worst-case test, the fitness function value for the hybrid algorithm was approximately 22% lower than that of the purely metaheuristic-based approach. Experimental tests further demonstrated that the hybrid algorithm, whether based on Hooke–Jeeves or Nelder–Mead, was stable and well suited for solving the considered problem. The article includes experimental results that confirm the effectiveness, accuracy, and efficiency of the proposed method. Full article

Deterministic lateral displacement (DLD) is a passive particle separation method based on microfluidic technology, with its separation mechanism primarily relying on particle size differences. Therefore, the critical separation size is of great significance in the design of DLD devices. The geometric asymmetry of the pillar array design significantly influences fluid behavior and critical particle size variations. This study first analyzed particle motion characteristics through particle trajectory observation experiments within asymmetric microfluidic chips. Subsequently, a two-dimensional numerical simulation method was employed to investigate the effects of three different ratios of lateral gap size to downstream gap size (G_x:G_y) on particle trajectories and flow field distribution. The results indicate that as G_x:G_y decreases, the upward flow rate gradually reduces, accompanied by changes in the flow field velocity distribution, causing particles to favor displacement mode. This study provides new theoretical foundations for the precise regulation of particle motion behavior and introduces novel insights for optimizing DLD device design. Full article

Microwave tomography serves as a promising non-invasive technique for breast imaging, yet accurate reconstruction in noisy environments remains challenging. We propose an adaptive subspace-based two-step iterative shrinkage/thresholding (S-TwIST) algorithm that enhances reconstruction accuracy through two key innovations: a singular value decomposition (SVD) approach for extracting deterministic contrast sources, and an adaptive strategy for optimal singular value selection. Unlike conventional DBIM methods that rely solely on secondary incident fields, S-TwIST incorporates deterministic induced currents to achieve more accurate total field approximation. The algorithm’s performance is validated using both synthetic “Austria” profiles and 45 digital breast phantoms derived from the UWCEM repository. The results demonstrate robust reconstruction capabilities across varying noise levels (0–20 dB SNR), achieving average relative errors of 0.4847% in breast tissue reconstruction without requiring prior noise level knowledge. The algorithm successfully recovers complex tissue structures and density distributions, showing potential for clinical breast imaging applications. Full article

The strategic planning of open-pit mining projects is highly influenced by geological variability, economic fluctuations, and operational uncertainties. Traditional deterministic optimization models often fail to account for these uncertainties, leading to potentially misleading economic evaluations. This paper proposes a hybrid model that integrates deterministic and stochastic optimization methods, following the principles of the ISO 31000 risk management framework, to comprehensively quantify uncertainty through key parameters affecting strategic mine planning. Monte Carlo simulations were applied to generate probability distributions of input parameters, including metal prices, mining and processing costs, and ore grade variability, allowing for a more robust financial assessment. The results demonstrate that while the deterministic approach estimates an NPV of USD 130.8 million, the stochastic model yields an average NPV of USD 155.5 million with a standard deviation of USD 76.5 million, highlighting the significant variability in financial outcomes. Risk assessment using Value at Risk (VaR) and Conditional Value at Risk (CVaR) further quantifies potential financial losses, revealing a 3% probability of project unprofitability. The developed methodology provides a structured approach to integrating uncertainty into mine planning, enabling more reliable economic evaluations and improving decision-making in strategic mining operations. Full article

Background: Understanding the dynamics of HIV transmission in heterogeneous populations is crucial for effective prevention strategies. This study introduces the Risk Modulation Point (RMP), a novel threshold identifying where HIV transmission transitions from unsustainable spread to self-sustaining epidemic dynamics. Methods: Using a deterministic, risk-stratified compartmental model, we examined HIV transmission across populations stratified into 100–200 risk groups, each characterized by behavioral heterogeneity modeled through a power-law distribution. The model captures key features of HIV progression, with simulations conducted across high- (~20%), moderate- (~5%), and low (~0.2%)-prevalence regimes. Results: Our findings reveal universal patterns in HIV dynamics. The RMP marks a consistent threshold across scenarios, separating low-risk groups where transmission is minimal from higher-risk groups sustaining the epidemic. Logistic growth in HIV prevalence across risk groups, with sharp transitions near the RMP, was observed universally. The force of infection follows power-law scaling, directly reflecting the level and nature of risk behavior within each group. Importantly, the location of the RMP remains largely invariant to the underlying sexual risk distribution, population resolution, and mixing patterns, making it applicable across both generalized and concentrated epidemics. Conclusion: The RMP framework offers actionable public health insights. It identifies key populations and transition regions for targeted interventions such as antiretroviral therapy and pre-exposure prophylaxis. By tracking shifts in the RMP, it also serves as an early warning indicator for epidemic transitions, guiding resource allocation and monitoring. The focus of the model on intrinsic epidemic dynamics, excluding external interventions, highlights its utility in uncovering fundamental transmission patterns. This study bridges theoretical modeling and practical application, providing a flexible framework for understanding HIV and other stratified epidemics. The findings advance HIV modeling by revealing generic patterns that transcend specific contexts, supporting data-driven public health strategies. Full article

We develop a deep reinforcement learning (RL) framework for an optimal market-making (MM) trading problem, specifically focusing on price processes with semi-Markov and Hawkes Jump-Diffusion dynamics. We begin by discussing the basics of RL and the deep RL framework used; we deployed the state-of-the-art Soft Actor–Critic (SAC) algorithm for the deep learning part. The SAC algorithm is an off-policy entropy maximization algorithm more suitable for tackling complex, high-dimensional problems with continuous state and action spaces, like those in optimal market-making (MM). We introduce the optimal MM problem considered, where we detail all the deterministic and stochastic processes that go into setting up an environment to simulate this strategy. Here, we also provide an in-depth overview of the jump-diffusion pricing dynamics used and our method for dealing with adverse selection within the limit order book, and we highlight the working parts of our optimization problem. Next, we discuss the training and testing results, where we provide visuals of how important deterministic and stochastic processes such as the bid/ask prices, trade executions, inventory, and the reward function evolved. Our study includes an analysis of simulated and real data. We include a discussion on the limitations of these results, which are important points for most diffusion style models in this setting. Full article