Search Results (56)

Existing nighttime visibility enhancement methods primarily focus on improving overall brightness under low-light conditions. However, nighttime road images are also affected by glare, glow, and flare from complex light sources such as streetlights and headlights, making it challenging to suppress locally overexposed regions and recover fine details. To address these challenges, we propose a Nighttime Road Glare Suppression Network (NRGS-Net) for glare removal and detail restoration. Specifically, to handle diverse glare disturbances caused by the uncertainty in light source positions and shapes, we designed a gated positional attention (GPA) module that integrates positional encoding with local contextual information to guide the network in accurately locating and suppressing glare regions, thereby enhancing the visibility of affected areas. Furthermore, we introduced an improved Uformer backbone named LCAtransformer, in which the downsampling layers adopt efficient depthwise separable convolutions to reduce computational cost while preserving critical spatial information. The upsampling layers incorporate a residual PixelShuffle module to achieve effective restoration in glare-affected regions. Additionally, channel attention is introduced within the Local Context-Aware Feed-Forward Network (LCA-FFN) to enable adaptive adjustment of feature weights, effectively suppressing irrelevant and interfering features. To advance the research in nighttime glare suppression, we constructed and publicly released the Night Road Glare Dataset (NRGD) captured in real nighttime road scenarios, enriching the evaluation system for this task. Experiments conducted on the Flare7K++ and NRGD, using five evaluation metrics and comparing six state-of-the-art methods, demonstrate that our method achieves superior performance in both subjective and objective metrics compared to existing advanced methods. Full article

The Mayanghe National Nature Reserve, a key habitat for the endangered François’ langur (Trachypithecus francoisi), faces significant anthropogenic disturbances, including extensive distribution of croplands, roads, and settlements. These human-modified features are predominantly concentrated at elevations between 500 and 800 m and on slopes of 10–20°, which notably overlap with the core elevation range utilized by François’ langur. Spatial analysis revealed that langurs primarily occupy areas within the 500–800 m elevation band, which comprises only 33% of the reserve but hosts a high density of human infrastructure—including approximately 4468 residential buildings and the majority of cropland and road networks. Despite slopes >60° representing just 18.52% of the area, langur habitat utilization peaked in these steep regions (exceeding 85.71%), indicating a strong preference for rugged karst terrain, likely due to reduced human interference. Habitat type analysis showed a clear preference for evergreen broadleaf forests (covering 37.19% of utilized areas), followed by shrublands. Landscape pattern metrics revealed high habitat fragmentation, with 457 discrete habitat patches and broadleaf forests displaying the highest edge density and total edge length. Connectivity analyses indicated that distribution areas exhibit a more continuous and aggregated habitat configuration than control areas. These results underscore François’ langur’s reliance on steep, forested karst habitats and highlight the urgent need to mitigate human-induced fragmentation in key elevation and slope zones to ensure the species’ long-term survival. Full article

This paper presents a data-driven energy management scheme for fuel cell and battery electric vehicles, formulated as a constrained optimal control problem. The proposed method employs a co-state network trained using real-time measurements to estimate the control law without requiring prior knowledge of the system model or a complete dataset across the full operating domain. In contrast to conventional reinforcement learning approaches, this method avoids the issue of high dimensionality and does not depend on extensive offline training. Robustness is demonstrated by treating uncertain and time-varying elements, including power consumption from air conditioning systems, variations in road slope, and passenger-related demands, as unknown disturbances. The desired state of charge is defined as a reference trajectory, and the control input is computed while ensuring compliance with all operational constraints. Validation results based on a combined driving profile confirm the effectiveness of the proposed controller in maintaining the battery charge, reducing fluctuations in fuel cell power output, and ensuring reliable performance under practical conditions. Comparative evaluations are conducted against two benchmark controllers: one designed to maintain a constant state of charge and another based on a soft actor–critic learning algorithm. Full article

Xylosalsola chiwensis (Popov) Akhani & Roalson is listed in the Red Data Book of Kazakhstan as a rare species with a limited distribution, occurring in small populations in Kazakhstan, Uzbekistan, and Turkmenistan. The aim of this study is to deepen the understanding of the ecological conditions of its habitats, the floristic composition of its associated plant communities, the species’ morphological and anatomical characteristics, and its molecular phylogeny, as well as to identify the main threats to its survival. The ecological conditions of the X. chiwensis habitats include coastal sandy plains and the slopes of chinks and denudation plains with gray–brown desert soils and bozyngens on the Mangyshlak Peninsula and the Ustyurt Plateau at altitudes ranging from −3 to 270 m above sea level. The species is capable of surviving in arid conditions (less than 100 mm of annual precipitation) and under extreme temperatures (air temperatures exceeding 45 °C and soil surface temperatures above 65 °C). In X. chiwensis communities, we recorded 53 species of vascular plants. Anthropogenic factors associated with livestock grazing, industrial disturbances, and off-road vehicle traffic along an unregulated network of dirt roads have been identified as contributing to population decline and the potential extinction of the species under conditions of unsustainable land use. The morphometric traits of X. chiwensis could be used for taxonomic analysis and for identifying diagnostic morphological characteristics to distinguish between species of Xylosalsola. The most taxonomically valuable characteristics include the fruit diameter (with wings) and the cone-shaped structure length, as they differ consistently between species and exhibit relatively low variability. Anatomical adaptations to arid conditions were observed, including a well-developed hypodermis, which is indicative of a water-conserving strategy. The moderate photosynthetic activity, reflected by a thinner palisade mesophyll layer, may be associated with reduced photosynthetic intensity, which is compensated for through structural mechanisms for water conservation. The flow cytometry analysis revealed a genome size of 2.483 ± 0.191 pg (2n/4x = 18), and the phylogenetic analysis confirmed the placement of X. chiwensis within the tribe Salsoleae of the subfamily Salsoloideae, supporting its taxonomic distinctness. To support the conservation of this rare species, measures are proposed to expand the area of the Ustyurt Nature Reserve through the establishment of cluster sites. Full article

This paper proposes a novel navigation approach for Unmanned Tracked Vehicles (UTVs) using prior-based ensemble deep reinforcement learning, which fuses the policy of the ensemble Deep Reinforcement Learning (DRL) and Dynamic Window Approach (DWA) to enhance both exploration efficiency and deployment safety in unstructured off-road environments. First, by integrating kinematic analysis, we introduce a novel state and an action space that account for rugged terrain features and track–ground interactions. Local elevation information and vehicle pose changes over consecutive time steps are used as inputs to the DRL model, enabling the UTVs to implicitly learn policies for safe navigation in complex terrains while minimizing the impact of slipping disturbances. Then, we introduce an ensemble Soft Actor–Critic (SAC) learning framework, which introduces the DWA as a behavioral prior, referred to as the SAC-based Hybrid Policy (SAC-HP). Ensemble SAC uses multiple policy networks to effectively reduce the variance of DRL outputs. We combine the DRL actions with the DWA method by reconstructing the hybrid Gaussian distribution of both. Experimental results indicate that the proposed SAC-HP converges faster than traditional SAC models, which enables efficient large-scale navigation tasks. Additionally, a penalty term in the reward function about energy optimization is proposed to reduce velocity oscillations, ensuring fast convergence and smooth robot movement. Scenarios with obstacles and rugged terrain have been considered to prove the SAC-HP’s efficiency, robustness, and smoothness when compared with the state of the art. Full article

There are significant emissions of greenhouse gases into the atmosphere from the transportation industry. As a result, the idea that electric vehicles (EVs) offer a revolutionary way to reduce greenhouse gas emissions and our reliance on rapidly depleting petroleum supplies has been put forward. EVs are becoming more common in many nations worldwide, and the rapid uptake of this technology is heavily reliant on the growth of charging stations. This is leading to a significant increase in their number on the road. This rise has created an opportunity for EVs to be integrated with the power system as a Demand Response (DR) resource in the form of an EV fast charging station (EVFCS). To allocate electric vehicle fast charging stations as a dynamic load for frequency control and on specific buses, this study included the optimal location for the EVFCS and the best controller selection to obtain the best outcomes as DR for various network disruptions. The optimal location for the EVFCS is determined by applying transient voltage drop and frequency nadir parameters to the Particle Swarm Optimization (PSO) location model as the first stage of this study. The second stage is to explore the optimal regulation of the dynamic EVFCS load using the PSO approach for the PID controller. PID controller settings are acquired to efficiently support power system stability in the event of disruptions. The suggested model addresses various types of system disturbances—generation reduction, load reduction, and line faults—when it comes to the Kundur Power System and the IEEE 39 bus system. The results show that Bus 1 then Bus 4 of the Kundur System and Bus 39 then Bus 1 in the IEEE 39 bus system are the best locations for dynamic EVFCS. Full article

A road network is an important spatial carrier for the efficient and reliable operation of urban services and material flows. In recent years, the “high road density, small block size” trend has become a major focus in urban planning practices. However, whether high-density road networks are highly resilient lacks quantitative evidence. This study presents a multi-scale analytical framework for measuring road network resilience from a topological perspective. We abstract 186 ideal orthogonal grid density models from an actual urban road network, quantifying resilience under two disturbance scenarios: random failures and intentional attacks. The results indicate that road network density indeed has a significant impact on resilience, with both scenarios showing a trend where higher densities correlate with greater resilience. However, the increase in resilience value under the intentional attack scenario is significantly higher than that under the random failure scenario. The findings indicate that network density plays a decisive role in determining resilience levels when critical edges fail. This is attributed to the greater presence of loops in denser networks, which helps maintain connectivity even under intentional disruption. In the random failure scenario, network resilience depends on the combined effects of the node degree and density. This study offers quantitative insights into the design of resilient urban forms in the face of disruptive events, establishing reference benchmarks for road network spacing at both meso- and micro-scales. The results provide practical guidance for resilient city planning in both newly developed and existing urban areas, supporting informed decision-making in urban morphology and disaster risk management. Full article

The research object of this paper is a new type of multi-functional, air-gap-type, vehicle-mounted magnetic suspension flywheel battery. It is a new energy storage technology with a long working life, high energy conversion efficiency, multiple charging and discharging times, low carbon and environmental protection. However, when the vehicle-mounted flywheel battery is operating, it will inevitably be disturbed by road conditions, resulting in loose sensors and feedback errors, thereby reducing the control accuracy and reliability of the system. To solve these problems, a sensorless control system came into being. It samples the current of the magnetic bearing coil through the hardware circuit and converts it into displacement for real-time control, eliminating the risk of sensor failure. However, the control accuracy of the traditional sensorless system is relatively low. Therefore, this paper adopts a BP (backpropagation) neural network PID controller based on genetic algorithm optimization on the basis of the sensorless control system. Through the joint simulation of the dynamic simulation software ADAMS/VIEW2018 and MATLAB2022b, the optimal PID control parameter database for complex road conditions is established. Through sensorless technology, the current of the flywheel battery is converted into the position error for extensive training so that the genetic BP neural network PID controller can accurately identify the current complex road conditions according to the position error, so as to provide the optimal PID control parameters corresponding to the road conditions to carry out accurate real-time stability control of the flywheel rotor. The experimental results show that the method can effectively reduce feedback errors, improve the control accuracy, and output optimal control parameters in real time under complex road conditions, which significantly improves the reliability and control performance of the vehicle flywheel battery system. Full article

Traffic network resilience refers to the ability of a traffic network to maintain a certain capacity and service level even when disturbed by external factors, as well as its capacity to recover following a disruptive event. This paper integrates traffic simulation with resilience analysis of urban road traffic networks and proposes a framework for identifying critical roads in urban road traffic networks from a resilience perspective. This framework is both theoretical and applicable to any unanticipated disruptive event. By incorporating four attribute indicators—traffic, topology, urban function, and socio-economic factors—the framework assesses the importance ranking of each road segment in an urban road traffic network both before and after an unanticipated disruptive event. A case study is conducted using a real urban road traffic network in Shanghai. From the perspective of policymakers, corresponding policy recommendations are made to enhance the resilience of urban road traffic networks against unanticipated disruptive events and to mitigate socio-economic losses. Full article

Bolstering the resilience of shrinking cities (SCs) is essential for maintaining urban dynamic security and fostering sustainable development. Accurately assessing and revealing the resilience level and impact mechanism of SCs to cope with disturbances and shocks has become a hot topic of research in urban sustainable development. In this research, we presented a systematic conceptualization of the fundamental components of urban shrinkage (US) and urban resilience (UR) and, based on US and UR theories, constructed a methodological framework aimed at investigating the spatiotemporal evolution mechanism and spatial correlation network effect of resilience in different SCs in China. This paper initially evaluates the UR levels of various types of SCs in China by establishing an evaluation model for US and a multidimensional evaluation index system for UR based on the theoretical frameworks, aligned with the national context in China. We also define the spatiotemporal evolution patterns of UR for different types of SCs. Subsequently, this paper employs a coupled coordination model and a geographical detector model to elucidate the influencing mechanisms on UR of different types of SCs, focusing on UR subsystems and indicators. Finally, this paper empirically examines the spatial correlation network effects of UR under various US scenarios using a social network analysis model. The results show that many SCs have progressively adjusted to the challenges posed by US, and the UR of SCs has shown an upward trend from 2010 to 2021. Cities with higher US levels generally show lower coordination in UR subsystems. The comprehensive utilization rate of industrial solid waste and road freight per capita are crucial for improving the UR of cities with higher US levels. Moreover, US probably strengthens UR connections between cities, facilitating resilience transmission and dissemination. These findings advance UR research within the US framework and offer theoretical foundations and planning guidance for environmentally friendly and high-quality development in shrinking cities. Full article

The rapid proliferation of vehicles globally presents significant challenges to road transportation efficiency and safety, including accidents, emissions, energy utilization, and road management. Autonomous vehicle platooning emerges as a promising solution within intelligent transportation systems, offering benefits like reduced fuel consumption and emissions, and optimized road use. However, implementing autonomous vehicle platooning faces obstacles such as stability under disturbances, safety protocols, communication networks, and precise control. This paper proposes a novel control strategy coordinated Kalman observer–Linear Quadratic Regulator (CKO-LQR) to ensure platoon formation stability in the presence of disturbances. The disturbances considered include vehicle movements, sensor noise, and communication delays, with the leading vehicle’s movement serving as the commanding signal. The proposed controller maintains a constant inter-gap distance between vehicles despite the disturbances utilizing a coordinated Kalman observer to estimate preceding vehicle movements. A comparative analysis with conventional PID controllers demonstrates superior performance in terms of faster settling times and robustness against disturbances. This research contributes to enhancing the efficiency and safety of autonomous vehicle platooning systems. Full article

A crucial role is played by steering-angle prediction in the control of autonomous vehicles (AVs). It mainly includes the prediction and control of the steering angle. However, the prediction accuracy and calculation efficiency of traditional YOLOv5 are limited. For the control of the steering angle, angular velocity is difficult to measure, and the angle control effect is affected by external disturbances and unknown friction. This paper proposes a lightweight steering angle prediction network model called YOLOv5Ms, based on YOLOv5, aiming to achieve accurate prediction while enhancing computational efficiency. Additionally, an adaptive output feedback control scheme with output constraints based on neural networks is proposed to regulate the predicted steering angle using the YOLOv5Ms algorithm effectively. Firstly, given that most lane-line data sets consist of simulated images and lack diversity, a novel lane data set derived from real roads is manually created to train the proposed network model. To improve real-time accuracy in steering-angle prediction and enhance effectiveness in steering control, we update the bounding box regression loss function with the generalized intersection over union (GIoU) to Shape-IoU_Loss as a better-converging regression loss function for bounding-box improvement. The YOLOv5Ms model achieves a 30.34% reduction in weight storage space while simultaneously improving accuracy by 7.38% compared to the YOLOv5s model. Furthermore, an adaptive output feedback control scheme with output constraints based on neural networks is introduced to regulate the predicted steering angle via YOLOv5Ms effectively. Moreover, utilizing the backstepping control method and introducing the Lyapunov barrier function enables us to design an adaptive neural network output feedback controller with output constraints. Finally, a strict stability analysis based on Lyapunov stability theory ensures the boundedness of all signals within the closed-loop system. Numerical simulations and experiments have shown that the proposed method provides a 39.16% better root mean squared error (RMSE) score than traditional backstepping control, and it achieves good estimation performance for angles, angular velocity, and unknown disturbances. Full article

This paper investigates a parameter-free H_∞ differential game approach for nonlinear active vehicle suspensions. The study accounts for the geometric nonlinearity of the half-car active suspension and the cubic nonlinearity of the damping elements. The nonlinear H_∞ control problem is reformulated as a zero-sum game between two players, leading to the establishment of the Hamilton–Jacobi–Isaacs (HJI) equation with a Nash equilibrium solution. To minimize reliance on model parameters during the solution process, an actor–critic framework employing neural networks is utilized to approximate the control policy and value function. An off-policy reinforcement learning method is implemented to iteratively solve the HJI equation. In this approach, the disturbance policy is derived directly from the value function, requiring only a limited amount of driving data to approximate the HJI equation’s solution. The primary innovation of this method lies in its capacity to effectively address system nonlinearities without the need for model parameters, making it particularly advantageous for practical engineering applications. Numerical simulations confirm the method’s effectiveness and applicable range. The off-policy reinforcement learning approach ensures the safety of the design process. For low-frequency road disturbances, the designed H_∞ control policy enhances both ride comfort and stability. Full article

The complex influence of various disturbances on an inertially stabilized platform (ISP) restricts the further improvement of its servo performance. This article investigates the mapping relationship between internal and external disturbances and servo performance by establishing a high-precision dynamics model of the servo device with simulation and experiment. For internal disturbances, a nonlinear model of friction and backlash is established based on a BP neural network, and the transmission error is reconstructed based on the principle of main order invariance. For external disturbances, the road disturbance torque, changing inertia, and mass imbalance torque are modeled. The quantitative mapping relationship between internal and external disturbances and servo performance is obtained through simulation, in which friction and road disturbance are the largest internal and external factors affecting the servo performance, respectively. These conclusions are verified by load simulation experiments on a certain type of servo device, in which the servo performance is improved by 17% for every 25% reduction of friction torque, and the servo performance is reduced by 12% for every 33% increase of road disturbance torque. The research results provide a reference for servo device selection, performance indicator assignment, and performance prediction of the ISP. Full article

Detecting vulnerable road users is a major challenge for autonomous vehicles due to their small size. Various sensor modalities have been investigated, including mono or stereo cameras and 3D LiDAR sensors, which are limited by environmental conditions and hardware costs. Radar sensors are a low-cost and robust option, with high-resolution 4D radar sensors being suitable for advanced detection tasks. However, they involve challenges such as few and irregularly distributed measurement points and disturbing artifacts. Learning-based approaches utilizing pillar-based networks show potential in overcoming these challenges. However, the severe sparsity of radar data makes detecting small objects with only a few points difficult. We extend a pillar network with our novel Sparsity-Robust Feature Fusion (SRFF) neck, which combines high- and low-level multi-resolution features through a lightweight attention mechanism. While low-level features aid in better localization, high-level features allow for better classification. As sparse input data are propagated through a network, the increasing effective receptive field leads to feature maps of different sparsities. The combination of features with different sparsities improves the robustness of the network for classes with few points. Full article