Poor Adaptation Research Articles

The impact of the COVID-19 lockdown on social and economic welfare in Uganda

BackgroundAs a measure to slow down the transmission of Coronavirus disease (COVID-19), governments around the world placed their countries under various stringent lockdown measures. Uganda is one of the countries that had a strict lockdown in Africa. This qualitative study explored the social and economic impacts of the COVID-19 lockdown in both an urban (Kampala) and rural (Wakiso) setting in Central Uganda.MethodsThe study used focus group discussions (FGDs), household interviews, and key informant interviews (KIIs). 14 FGDs were conducted among several stakeholders including community health workers, health professionals, and members of the community. 40 household interviews were conducted among low, middle, and high-income households, while 31 KIIs were held among policy makers, non-governmental organisations, and the private sector. Data were analysed thematically in NVivo 2020 (QSR International).ResultsFindings from the study are presented under six themes: family disruption; abuse of children’s rights; disruption in education; food insecurity; impact on livelihoods; and violation of human rights. The study found that the COVID-19 lockdown led to family breakups, loss of family housing, as well as increased both caring responsibilities and gender-based violence especially towards females. Children’s welfare suffered through increased child labour, sexual exploitation, and early marriages. The extended closure of schools led to delayed educational milestones, poor adaptation to home-based learning, and increased school drop-out rates. Increased food insecurity led to changes in feeding patterns and reduced food varieties. Livelihoods were negatively affected hence people depleted their savings and capital. Unlawful detention and beating by law enforcement officers increased during the lockdown.ConclusionFuture pandemic planning needs to consider the consequences of lockdown on the social and economic wellbeing of communities hence put in place appropriate mitigation measures during and after the outbreak.

Study on energy-saving applications of vacuum ejectors across various operational demands

Vacuum suction technology is a pivotal innovation within the Industry 4.0 revolution. Ejectors serve as the core of vacuum systems with distinct advantages such as simple structure, low cost, and ease of operation. However, traditional fixed ejectors suffer from low efficiency, energy waste, and poor adaptability. This study aims to optimize the design of vacuum ejectors to meet various requirements, achieving energy efficiency and high performance in automated production. On this basis, an energy-saving vacuum system based on dual ejectors is proposed, which can switch between single and dual ejectors depending on different working stages. When the vacuum reaches 90 kPa, the required motive pressure and the entrainment ratio are selected as the screening indicators for the vacuum ejector. A two-dimensional axisymmetric model is established and verified through experimentation. The key parameters are identified through a single-factor analysis. A multi-factor analysis is employed to determine the recommended values for each parameter and their respective applications. The energy-saving effects of the optimized ejectors are evaluated in conjunction with the automated processes. The findings indicate that the vacuum system based on dual ejectors can achieve rapid response and low air consumption. This paper contributes to advancing the research on energy-saving applications of ejectors.

Growth and Fecundity of Nonnative Blue Honeysuckle Cultivars: Comparisons with Native and Invasive Congeners in a Maine Field Trial

Blue honeysuckle (Lonicera caerulea) is a circumpolar species complex with representatives in Europe, Asia, and North America. Although honeysuckles (Lonicera spp.) from Eurasia have a history of invasiveness in North America, farmers and homeowners are interested in growing nonnative blue honeysuckle hybrids because of their edible blue fruits. To assess whether these cultivars and closely related native blue honeysuckles (Lonicera caerulea subsp. villosa) might have similar growth and fecundity, we planted five nonnative cultivars of blue honeysuckle and five native genotypes in a common garden in Orono, ME, USA, along with invasive red-fruited honeysuckles [Tatarian honeysuckle (Lonicera tatarica) and European fly honeysuckle (Lonicera xylosteum)] for comparison. Rooted cuttings were planted into a field plot in Jun 2016 and fully maintained during the first season; thereafter, maintenance consisted of weeding once annually. Seventy-three percent of native blue honeysuckle plants survived to the end of the study, whereas survival and growth of nonnative cultivars were more robust. In 2021, nonnative cultivars had an average height of 81 cm and width of 86 cm, which were 2.8 times the height and 2.9 times the width of surviving native plants. The estimated canopy volumes of nonnative blue honeysuckles were an average of 20 times those of their native counterparts. The bloom periods of native and nonnative blue honeysuckles overlapped considerably. However, only seven of the 22 living native plants produced fruits in 2021, with an average of three fruits per plant among them. In contrast, nearly all plants of the nonnative cultivars produced fruits, with an average of 616 fruits per plant. In comparison, the red-fruited invasives had an average of 9739 fruits per plant. Native blue honeysuckles produced very few seeds, whereas nonnative cultivars had an average of 13,918 seeds per plant, which was approximately one-fourth the number produced by invasive red-fruited honeysuckles. We concluded that native and nonnative genotypes of blue honeysuckle differ strikingly in survival, growth, and production of fruits and seeds. However, invasive red-fruited honeysuckles grew faster with higher fecundity than nonnative blue honeysuckles in our full-sun landscape. Because bloom times overlapped substantially between native and nonnative blue honeysuckles, the potential for gene flow to occur from planted cultivars into native populations merits consideration. Several possible explanations of differences in performance among blue honeysuckles include hybrid vigor of cultivars or shallow rooting or poor adaptability of native genotypes to the environment of the common-garden trial. Our results, which demonstrated that nonnative blue honeysuckles are likely to be distinct from their native relatives in terms of competitiveness and fecundity, suggest that caution is warranted during the introduction and cultivation of agricultural genotypes.

‘Breakdown and healing’ - adaptation experiences of postpartum nurses returning to work: a descriptive phenomenological study

BackgroundPromoting the work adaptation of nurses returning to work after childbirth is key to improving their physical and mental health and ensuring the quality of nursing care. Although postpartum nurses have reported high work stress and poor adaptation after returning to work, little is known regarding the experiences, problems and further needs. This study aimed to explore the adaptation experiences of postpartum nurses returning to work.MethodsThis was a qualitative study with descriptive phenomenological approach. Individual semi-structured interviews were conducted between March and May 2023 in four tertiary hospitals in Changsha, mainland China, which were selected using convenience sampling. A total of 16 postpartum nurses returning to work within one year were recruited using purposive sampling and maximum variation sampling. The interviews were analysed using the Colaizzi’s 7-step method.Results‘Breakdown and healing’ is a strong thread in postpartum nurses’ work adaptation. It links 13 subthemes in the interview data, which can be grouped under three overarching themes: (a) changes and challenges of multiple roles; (b) self-coping and social support; and (c) further needs after returning to work. The phrase ‘breakdown and healing’ reflects the mutual relationship between stress and coping among postpartum nurses. The changes and challenges associated with being required to fulfil multiple roles of mothers, wives and nurses make the subjects feel on the verge of ‘breakdown’. When returning to work, most of them move slowly towards ‘healing’ through positive self-management and support from their partners, families, supervisors and colleagues. Moreover, their ongoing demands and unsolved problems – such as their desire for a workplace that is friendly and serves their needs – were generated by the constant process of ‘breakdown and healing’.ConclusionsUnderstanding how postpartum nurses returning to work perceive adaptation experiences is essential for nursing managers, who can use the findings to implement targeted measures to shorten postpartum nurses’ maladaptation period after returning to work and promote their work adaptation. This study underscores the critical importance of personalized return-to-work training, flexible work arrangements, support from colleagues and managers, well-established maternal facilities and services, and targeted policy efforts in enhancing postpartum nurses’ work adaptation.

Challenges and Teaching Reform Strategies in Vocational College Engineering Training Courses

With the rapid development of vocational education, engineering training has become a crucial course for cultivating highly skilled technical professionals, highlighting its importance. However, vocational college engineering training courses still face numerous challenges during the actual teaching process, which affect the effectiveness of course teaching and the enhancement of students’ engineering competence. This article thoroughly analyzes the main problems existing in current engineering training course teaching, focusing on four aspects: students’ cognitive differences, inadequate course development, poor course adaptability, and limited teaching resources. Based on this analysis and previous research findings, targeted strategies are proposed, including enhancing students’ course cognition, clarifying course positioning, improving course adaptability, and optimizing the allocation of teaching resources. Furthermore, this article preliminarily verifies the effectiveness of some reform strategies through a small-scale teaching experiment. This study aims to provide theoretical guidance and practical reference for enhancing the teaching quality of vocational college engineering training courses.

A DNA-Like blockchain model: implementing decentralized management, dynamic scheme adjustment, and data security for optimal service composition

ABSTRACT Cloud manufacturing (CM) promotes the development of small and medium-sized manufacturing enterprises (SMMEs) by integrating resources and collaboratively providing users with diversified services. However, the implementation of CM for SMMEs faces several challenges, such as the excessive centralization of enterprise management, poor dynamic adaptability of service composition, and insecurity of data information. To address these issues, this paper proposes a DNA-like blockchain model (DNABM) based on blockchain technology. The DNABM consists of a service information chain (SIC) and a manufacturing process chain (MPC). The SIC nodes (SIblocks) are formed based on the corresponding subtasks to record information about all manufacturing services that can complete the subtasks. The MPC nodes (MPblocks) are formed by selecting the most suitable services from the SIblocks. The DNABM has advantages of flexible service management, dynamically adaptable service compositions, and securely managed manufacturing enterprise data across various levels. Finally, taking the foundry industry as an example, demonstrating how the DNABM can promote the collaboration of SMMEs in CM. However, the actual implementation of the DNABM is still limited by the large number of heterogeneous tasks and services, the low utilization rate of process data, and the lack of coordination compatibility between services.

An Early Warning Model for Turbine Intermediate-Stage Flux Failure Based on an Improved HEOA Algorithm Optimizing DMSE-GRU Model

As renewable energy sources such as wind and photovoltaics continue to enter the grid, their intermittency and instability leads to an increasing demand for peaking and frequency regulation. An efficient dynamic monitoring method is necessary to improve the safety level of intelligent operation and maintenance of power stations. To overcome the insufficient detection accuracy and poor adaptability of traditional methods, a novel fault early warning method with careful consideration of dynamic characteristics and model optimization is proposed. A combined loss function is proposed based on the dynamic time warping and the mean square error from the perspective of both shape similarity and time similarity. A prediction model of steam turbine intermediate-stage extraction temperature based on the gate recurrent unit is then proposed, and the change in prediction residuals is utilized as a fault warning criterion. In order to further improve the diagnostic accuracy, a human evolutionary optimization algorithm with lens opposition-based learning is proposed for model parameter adaptive optimization. Experiments on real-world normal and faulty operational data demonstrate that the proposed method can improve the detection accuracy by an average of 1.31% and 1.03% compared to the long short-term memory network, convolutional neural network, back propagation network, extreme learning machines, gradient boosting decision tree, and LightGBM models.

Development of a wheeled wall‐climbing robot with an internal corner wall adaptive magnetic adhesion mechanism

AbstractThe wall‐climbing robot is a growing trend for robotized intelligent manufacturing of large and complex components in shipbuilding, petrochemical, and other industries, while several challenges remain to be solved, namely, low payload‐to‐weight ratio, poor surface adaptability, and ineffective traversal maneuverability, especially on noncontinuous surfaces with internal corners (non‐CSIC). This paper designs a high payload‐to‐weight ratio wheeled wall‐climbing robot which can travel non‐CSIC effectively with a payload capacity of up to 75 kg, and it can carry a maximum load of 141.5 kg on a vertical wall. By introducing a semi‐enclosed magnetic adhesion mechanism, the robot preserves a redundant magnetic adsorption ability despite the occurrence of significant gaps between localized body components and the wall surface. In addition, by ingeniously engineering a passive adaptive module into the robot, both the surface adaptability and crossability are enhanced without increasing the gap between the body and the wall, thereby ensuring the optimization of the adsorption force. Considering the payload capacity and diversity when climbing on vertical walls, inclined walls, ceilings, and internal corner transitions, control equations for internal corner transitions and comprehensive simulations of magnetic adsorption forces are performed using FEA tools. Finally, a functional prototype was developed for rigorous experimental testing, with the results confirming that the robot successfully meets the desired functionality and performance benchmarks.

Deep-sea microorganisms-driven V5+ and Cd2+ removal from vanadium smelting wastewater: Bacterial screening, performance and mechanism

The disorderly discharge of industrial wastewater containing heavy metals has caused serious water pollution and ecological environmental risks, ultimately threatening human life and health. Biological treatment methods have obvious advantages, but the existing microorganisms exhibit issues such as poor resistance, adaptability, colonization ability, and low activity. However, a wide variety of microorganisms in deep-sea hydrothermal vent areas are tolerant to heavy metals, possessing the potential for efficient treatment of heavy metal wastewater. Based on this, the study obtained a group of deep-sea microbial communities dominated by Burkholderia-Caballeronia-Paraburkholderia through shake flask experiments from the sediments of deep-sea hydrothermal vents, which can simultaneously achieve the synchronous removal of vanadium and cadmium heavy metals through bioreduction, biosorption, and biomineralization. Through SEM-EDS, XRD, XPS, and FT-IR analyses, it was found that V(V) was reduced to V(IV) through a reduction process and subsequently precipitated. Glucose oxidation accelerated this process. Cd(II) underwent biomineralization to form precipitates such as cadmium hydroxide and cadmium carbonate. Functional groups on the microbial cell surface, such as -CH2, C=O, N-H, -COOH, phosphate groups, amino groups, and M-O moieties, participated in the bioadsorption processes of V(V) and Cd(II) heavy metals. Under optimal conditions, namely a temperature of 40 °C, pH value of 7.5, inoculation amount of 10%, salinity of 4%, COD concentration of 600 mg/L, V5+ concentration of 300 mg/L, and Cd2+ concentration of 40 mg/L, the OD600 can reach its highest at 72 h, with the removal efficiency of V5+, Cd2+, and COD in simulated vanadium smelting wastewater reaching 86.32%, 59.13%, and 61.63%, respectively. This study provides theoretical insights and practical evidence for understanding the dynamic changes in microbial community structure under heavy metal stress, as well as the resistance mechanisms of microbial treatment of industrial heavy metal wastewater.

Fall Detection Method for Infrared Videos Based on Spatial-Temporal Graph Convolutional Network.

The timely detection of falls and alerting medical aid is critical for health monitoring in elderly individuals living alone. This paper mainly focuses on issues such as poor adaptability, privacy infringement, and low recognition accuracy associated with traditional visual sensor-based fall detection. We propose an infrared video-based fall detection method utilizing spatial-temporal graph convolutional networks (ST-GCNs) to address these challenges. Our method used fine-tuned AlphaPose to extract 2D human skeleton sequences from infrared videos. Subsequently, the skeleton data was represented in Cartesian and polar coordinates and processed through a two-stream ST-GCN to recognize fall behaviors promptly. To enhance the network's recognition capability for fall actions, we improved the adjacency matrix of graph convolutional units and introduced multi-scale temporal graph convolution units. To facilitate practical deployment, we optimized time window and network depth of the ST-GCN, striking a balance between model accuracy and speed. The experimental results on a proprietary infrared human action recognition dataset demonstrated that our proposed algorithm accurately identifies fall behaviors with the highest accuracy of 96%. Moreover, our algorithm performed robustly, identifying falls in both near-infrared and thermal-infrared videos.

Real-Time Estimation of Human Arm Dynamic Parameters Based on Optical

Abstract A flexible optical intensity sensor is designed in this study to address the poor adaptability of fixed-value parameters in exoskeleton research, which makes it difficult to handle system changes and uncertainties. The light transmission portion of the sensor is made of TPU and PET, exhibiting flexibility and bendable deformation characteristics, as well as good sensitivity, anti-interference capability, and stability. The flexible optical sensor is fixed on the side of the binding device, which is installed on the biceps brachii to collect the optical intensity loss signals after sensor bending. By utilizing this method, parameters such as the cross-sectional radius, arm centroid, and moment of inertia can be estimated in real-time. Finally, the estimated radius values are compared with theoretical values, and the accuracy is verified by combining the kinematic model. Experimental results demonstrate that the proposed method achieves a parameter information error of less than 8.63% and a stability better than 4.3%. Compared to the traditional fixed parameter method, the stability has improved by 25.06%, and the accuracy has improved by 9.04%.

Lightweight bobbin yarn detection model for auto-coner with yarn bank

The automated replacement of empty tubes in the yarn bank is a critical step in the process of automatic winding machines with yarn banks, as the real-time detection of depleted yarn on spools and accurate positioning of empty tubes directly impact the production efficiency of winding machines. Addressing the shortcomings of traditional methods, such as poor adaptability and low sensitivity in optical and visual tube detection, and aiming to reduce the computational and detection time costs introduced by neural networks, this paper proposes a lightweight yarn spool detection model based on YOLOv8. The model utilizes Darknet-53 as the backbone network, and due to the dense spatial distribution of yarn spool targets, it incorporates large selective kernel units to enhance the recognition and positioning of dense targets. To address the issue of excessive focus on local features by convolutional neural networks, a bi-level routing attention mechanism is introduced to capture long-distance dependencies dynamically. Furthermore, to balance accuracy and detection speed, a FasterNeck is constructed as the neck network, replacing the original convolutional blocks with Ghost convolutions and integrating with FasterNet. This design minimizes the sacrifice of detection accuracy while achieving a significant improvement in inference speed. Lastly, the model employs weighted IoU with a dynamic focusing mechanism as the bounding box loss function. Experimental results on a custom yarn spool dataset demonstrate a notable improvement over the baseline model, with a high-confidence mAP of 94.2% and a compact weight size of only 4.9 MB. The detection speed reaches 223FPS, meeting the requirements for industrial deployment and real-time detection.

COMPUTER PRODUCTION OF FACIAL EPITHETS

BACKGROUND: The issue of rehabilitation of patients with facial defects is extremely acute. In addition to the annual increase in the number of patients with cancer of the maxillofacial region, in recent years the number of people with shrapnel and gunshot wounds to the face has increased as a result of local wars and conflicts. Traditional methods of orthopedic rehabilitation of patients and the manufacture of facial epitheses are quite a complex and lengthy process. In the postoperative period, there is a sharp decrease in the quality of life of this category of patients, a violation of the basic body functions necessary for vital activity and poor social adaptation. Direct prosthetics of the face in the postoperative period was impossible due to the lack of necessary digital modeling technologies and structural materials for additive or subtractive production methods. The production of temporary or permanent facial epitheses using digital technologies is an urgent task that can improve the social and functional living conditions of patients. AIMS: Development of 3D modeling technology for additive manufacturing of facial epithets. METHODS: To achieve this goal, the first task was to develop specialized three-dimensional software for modeling defects in the facial area. The functionality of the program should allow you to virtually simulate the missing parts of the face (ear, eye, nose, orbit). To create a digital platform together with IT specialists, it was decided to use the following programming languages: C++ - Writing the software core, writing UI/UX interaction modules, interacting with the Windows operating system; C# - Complex assembly of the entire project; Python is an automated assembly of virtual library modules; OpenGL HSLS is a shader language for graphical visualization of objects; C is the creation of functions for interacting with shaders that require high speed. RESULTS: A specialized computer program has been developed for 3D modeling of prostheses in patients with defects in the middle zone of the face using combined facial scanning and computed tomography data (Computer program. Apresyan S.V., Stepanov A.G. "A program for 3D modeling of facial epitheses. Registration number (certificate) 2023663490, Registration date: 07/04/2023). Instead of obtaining analog impressions with plaster or silicone material, the developed technology uses a special 3D facial scanner, which greatly eases the suffering of patients. A virtual three-dimensional database of ears, noses, orbits, and zygomatic bones of patients of various ages and gender is integrated into the developed program. This allows the specialist to select the most adaptive part of the face to make up for the defect. Built-in modeling tools allow you to personalize a 3D model of a part of the face, based on the structural features of the maxillofacial region of a person. The finished three-dimensional model of a part of the face can be exported in various formats CONCLUSIONS: The developed 3D program for modeling defects allows to avoid invasive prosthetics approaches, to coordinate the shape of future structures with the patient. The built-in library of structures with a database provides remote manufacturing of the prosthesis without the presence of the patient, in case of a need to replace it. Among the undeniable advantages of the technology, it is important to include the fact that prostheses can be made directly on the day of surgery to remove part of the face, completely restoring lost functions and providing rapid social adaptation.

An improved infrared simulation method based on generative adversarial networks

Infrared simulation technology has great application value in the fields of UAV target simulation and infrared signal processing. However, traditional infrared simulation technology has the disadvantages of poor scene adaptability, low fidelity of infrared simulation images and long training time. This paper proposes an infrared image simulation algorithm based on improved Generative Adversarial Networks(GAN) to generate multi-scene infrared images, which could provide a reliable expansion of infrared database. First, we construct a GAN model by introducing a multi-resolution hierarchical structure and attention mechanism, which can effectively enhance the model’s adaptability to different scenes during the simulation of generating infrared images. Secondly, the SSIM loss term and L1 loss term are introduced during the training process to form a composite loss function, which prompts the proposed GAN to generate high-fidelity infrared simulation images. Finally, while maintaining the quality of simulated infrared images, a lightweight model is designed through model compression to significantly reduce the number of model parameters.The traditional discriminator based on the UNet-self attention is replaced by a discriminator based on attention mechanisms and adaptive max-pooling structure. The proposed GAN is trained and tested on three public datasets. Experimental results demonstrate that the proposed method improves structural similarity by 11% compared to Pix2Pix and achieves a Peak Signal-to-Noise Ratio (PSNR) of 25.8, which is 3.9 higher than InfraGAN. At the same time, compared with InfraGAN, the number of model parameters is reduced by 68% , which makes the method efficiently obtain simulated infrared images corresponding to given visible images.

Physiological response and molecular mechanism of glyphosate degradation by Pseudomonas alcaligenes Z1–1

The accumulation of glyphosate during its application in agriculture and its toxicity seriously threaten ecosystems and human health. Currently, glyphosate residual contamination is mainly accomplished through bioremediation techniques based on enhanced microbial degradation activity. However, there are drawbacks, such as poor environmental adaptability of strains and low degradation efficiency. Therefore, in this study, an efficient glyphosate-degrading strain, Pseudomonas alcaligenes Z1–1, was isolated from herbicide-contaminated environments and was capable of completely degrading glyphosate at a concentration of 200 mg/L within 7 days. Kinetics analysis showed that glyphosate degradation was concentration-dependent, with a maximum tolerant concentration of 800 mg/L. Mass spectrometric analysis indicated that AminoMethylPhosphonic acid (AMPA) was the predominant intermediate produced in the degradation pathway of glyphosate, revealing that glyphosate destruction began with breaking the C-N bond. Whole genome sequencing identified the key genes potentially involved in glyphosate degradation, including the thiO, glpA, aroA, soxB, and argA genes. Furthermore, in contrast to the majority of the metabolic pathways previously reported for glyphosate degradation via glyphosate oxidoreductase, the breaking of the C-N bond was primarily catalyzed by glycine oxidase. Overall, this research provides novel insights into the mechanisms of glyphosate degradation, offering valuable degradation enzyme resources for future applications.

Fusion of Multi-Layer Attention Mechanisms and CNN-LSTM for Fault Prediction in Marine Diesel Engines

Timely and effective maintenance is imperative to minimize operational disruptions and ensure the reliability of marine vessels. However, given the low early warning rates and poor adaptability under complex conditions of previous data-driven fault prediction methods, this paper presents a hybrid deep learning model based on multi-layer attention mechanisms for predicting faults in a marine diesel engine. Specifically, this hybrid model first introduces a Convolutional Neural Network (CNN) and self-attention to extract local features from multi-feature input sequences. Then, we utilize Long Short-Term Memory (LSTM) and multi-head attention to capture global correlations across time steps. Finally, the hybrid deep learning model is integrated with the Exponential Weighted Moving Average (EWMA) to monitor the operational status and predict potential faults in the marine diesel engine. We conducted extensive evaluations using real datasets under three operating conditions. The experimental results indicate that the proposed method outperforms the current state-of-the-art methods. Moreover, ablation studies and visualizations highlight the importance of fusing multi-layer attention, and the results under various operating conditions and application scenarios demonstrate that this method possesses predictive accuracy and broad applicability. Hence, this approach can provide decision support for condition monitoring and predictive maintenance of marine mechanical systems.

Development of a Highly Adaptive Miniature Piezoelectric Robot Inspired by Earthworms.

Miniature resonant piezoelectric robots have the advantages of compact structure, fast response, high speed, and easy control, which have attracted the interest of many scholars in recent years. However, piezoelectric robots usually suffer from the problem of poor adaptability due to the micron-level amplitude at the feet. Inspired by the fact that earthworms have actuation trajectories all around their bodies to move flexibly under the ground, a miniature piezoelectric robot with circumferentially arranged driving feet to improve adaptability is proposed. Notably, a longitudinal-vibration-compound actuation principle with multilegged collaboration is designed to achieve the actuation trajectories around the robot, similar to the earthworms. The structure and operating principle are simulated by the finite element method, and the prototype is fabricated. The robot weighs 22.7g and has dimensions of 35.5×36.5×47mm3. The robot is tethered to an ultrasonic power supply, and the experimental results show that the speed reaches 179.35mms-1 under an exciting signal with a frequency of 58.5kHz and a voltage of 200Vp-p. High adaptability is achieved by the proposed robot, it can move on flat, fold, concave, and convex surfaces, and even in an inclined or rotating tube.

A panoramic driving perception fusion algorithm based on multi-task learning.

With the rapid development of intelligent connected vehicles, there is an increasing demand for hardware facilities and onboard systems of driver assistance systems. Currently, most vehicles are constrained by the hardware resources of onboard systems, which mainly process single-task and single-sensor data. This poses a significant challenge in achieving complex panoramic driving perception technology. While the panoramic driving perception algorithm YOLOP has achieved outstanding performance in multi-task processing, it suffers from poor adaptability of feature map pooling operations and loss of details during downsampling. To address these issues, this paper proposes a panoramic driving perception fusion algorithm based on multi-task learning. The model training involves the introduction of different loss functions and a series of processing steps for lidar point cloud data. Subsequently, the perception information from lidar and vision sensors is fused to achieve synchronized processing of multi-task and multi-sensor data, thereby effectively improving the performance and reliability of the panoramic driving perception system. To evaluate the performance of the proposed algorithm in multi-task processing, the BDD100K dataset is used. The results demonstrate that, compared to the YOLOP model, the multi-task learning network performs better in lane detection, drivable area detection, and vehicle detection tasks. Specifically, the lane detection accuracy improves by 11.6%, the mean Intersection over Union (mIoU) for drivable area detection increases by 2.1%, and the mean Average Precision at 50% IoU (mAP50) for vehicle detection improves by 3.7%.

Fisher discrimination multiple kernel dictionary learning for robust identification of nonlinear features in machinery health monitoring

In sparse representation-based classification, Fisher discrimination dictionary learning (FDDL) has attracted widespread attention due to its advantages such as fewer parameters and good dictionary discriminability. However, due to its linear nature, it is difficult to handle nonlinear features. Kernel-based nonlinear transformation enables dictionary learning to capture nonlinear features embedded in samples, but traditional single-kernel methods have limited performance. In this paper, a Fisher discrimination multiple kernel dictionary learning (FDMKDL) method is proposed, in which Fisher discriminative criterion is imposed on both high-dimensional samples and coding coefficients. Specifically, we derive the kernel version of FDDL, i.e., Fisher discrimination kernel dictionary learning (FDKDL) to learn nonlinear features and promote dictionary discriminability. Meanwhile, to avoid the problem of poor adaptability and possible manual selection caused by single-kernel methods, we further derive a flexible multiple kernel learning (MKL) framework, which utilizes the complementary information of multiple kernel functions to adaptively obtain the optimal weights and synthesize a discriminative kernel space. Finally, FDMKDL combines FDKDL with this MKL framework to obtain a more discriminative dictionary. Experimental evaluations conducted on two datasets demonstrate the effectiveness and robustness of the proposed FDMKDL method in learning nonlinear features for machinery health monitoring when compared to several state-of-the-art methods.

Research on driving condition recognition in hybrid electric vehicle energy management strategy

Abstract The driving conditions encountered by vehicles during operation are complex and varied, and the ECMS (Equivalent fuel Consumption Minimization Strategy) based on minimum equivalent fuel consumption has poor adaptability to different driving conditions. To optimize the adaptability of the ECMS and improve the overall vehicle economy, this paper proposes an operating condition recognition model based on the Back Propagation neural network algorithm to optimize the ECMS energy management control strategy. Firstly, three types of driving conditions were identified: urban, suburban, and high-speed conditions. A complete vehicle model was constructed on the Matlab/Simulink platform. The efficacy of the strategy, before and after optimization, is verified under the WLTP (Worldwide Harmonized Light Vehicles Test Procedure) driving cycle. Simulation results show that the optimized ECMS can better adjust the power apportionment among the engine and motor, reducing the fuel consumption per hundred kilometers by 4.01% and thereby enhancing the overall vehicle economy.