Search results

Because of the various advantages of reinforcement learning (RL) mentioned above, this study uses RL to train unmanned aerial vehicles to perform two tasks: target search and cooperative obstacle avoidance.

This study draws inspiration from the recurrent state-space model and recurrent models (RPM) to propose a simpler yet highly effective model called the unmanned aerial vehicles prediction model (UAVPM). The main objective is to assist in training the UAV representation model with a recurrent neural network, using the soft actor-critic algorithm.

This study proposes a generalized actor-critic framework consisting of three modules: representation, policy and value. This architecture serves as the foundation for training UAVPM. This study proposes the UAVPM, which is designed to aid in training the recurrent representation using the transition model, reward recovery model and observation recovery model. Unlike traditional approaches reliant solely on reward signals, RPM incorporates temporal information. In addition, it allows the inclusion of extra knowledge or information from virtual training environments. This study designs UAV target search and UAV cooperative obstacle avoidance tasks. The algorithm outperforms baselines in these two environments.

It is important to note that UAVPM does not play a role in the inference phase. This means that the representation model and policy remain independent of UAVPM. Consequently, this study can introduce additional “cheating” information from virtual training environments to guide the UAV representation without concerns about its real-world existence. By leveraging historical information more effectively, this study enhances UAVs’ decision-making abilities, thus improving the performance of both tasks at hand.

This paper aims to study the agent learning from expert demonstration data while incorporating reinforcement learning (RL), which enables the agent to break through the limitations of expert demonstration data and reduces the dimensionality of the agent’s exploration space to speed up the training convergence rate.

Firstly, the decay weight function is set in the objective function of the agent’s training to combine both types of methods, and both RL and imitation learning (IL) are considered to guide the agent's behavior when updating the policy. Second, this study designs a coupling utilization method between the demonstration trajectory and the training experience, so that samples from both aspects can be combined during the agent’s learning process, and the utilization rate of the data and the agent’s learning speed can be improved.

The method is superior to other algorithms in terms of convergence speed and decision stability, avoiding training from scratch for reward values, and breaking through the restrictions brought by demonstration data.

The agent can adapt to dynamic scenes through exploration and trial-and-error mechanisms based on the experience of demonstrating trajectories. The demonstration data set used in IL and the experience samples obtained in the process of RL are coupled and used to improve the data utilization efficiency and the generalization ability of the agent.

Vehicle allocation problems (VAPs), which are frequently confronted in many transportation activities, primarily including but not limited to full truckload freight transportation operations, induce a significant economic impact. Despite the increasing academic attention to the field, literature still fails to match the needs of and opportunities in the growing industrial practices. In particular, the literature can grow upon the ideas on sustainability, Industry 4.0 and collaboration, which shape future practices not only in logistics but also in many other industries. This review has the potential to enhance and accelerate the development of relevant literature that matches the challenges confronted in industrial problems. Furthermore, this review can help to explore the existing methods, algorithms and techniques employed to address this problem, reveal directions and generate inspiration for potential improvements.

This study provides a literature review on VAPs, focusing on quantitative models that incorporate any of the following emerging logistics trends: sustainability, Industry 4.0 and logistics collaboration.

In the literature, sustainability interactions have been limited to environmental externalities (mostly reducing operational-level emissions) and economic considerations; however, emissions generated throughout the supply chain, other environmental externalities such as waste and product deterioration, or the level of stakeholder engagement, etc., are to be monitored in order to achieve overall climate-neutral services to the society. Moreover, even though there are many types of collaboration (such as co-opetition and vertical collaboration) and Industry 4.0 opportunities (such as sharing information and comanaging distribution operations) that could improve vehicle allocation operations, these topics have not yet received sufficient attention from researchers.

The scientific contribution of this study is twofold: (1) This study analyses decision models of each reviewed article in terms of decision variable, constraint and assumption sets, objectives, modeling and solving approaches, the contribution of the article and the way that any of sustainability, Industry 4.0 and collaboration aspects are incorporated into the model. (2) The authors provide a discussion on the gaps in the related literature, particularly focusing on practical opportunities and serving climate-neutrality targets, carried out under four main streams: logistics collaboration possibilities, supply chain risks, smart solutions and various other potential practices. As a result, the review provides several gaps in the literature and/or potential research ideas that can improve the literature and may provide positive industrial impacts, particularly on how logistics collaboration may be further engaged, which supply chain risks are to be incorporated into decision models, and how smart solutions can be employed to cope with uncertainty and improve the effectiveness and efficiency of operations.

This paper aims to study a neural network-based fault-tolerant controller to improve the tracking control performance of an unmanned autonomous helicopter with system uncertainty, external disturbances and sensor faults, using the prescribed performance method.

To ensure that the tracking error satisfies the prescribed performance, the authors adopt an error transformation function method. A control scheme based on the neural network and high-order disturbance observer is designed to guarantee the boundedness of the closed-loop system. A simulation is performed to prove the validity of the control scheme.

The developed adaptive fault-tolerant control method makes the system with sensor fault realize tracking control. The error transformation function method can effectively handle the prescribed performance requirements. Sensor fault can be regarded as a type of system uncertainty. The uncertainty can be approximated accurately using neural networks. A high-order disturbance observer can effectively suppress compound disturbances.

The tracking performance requirements of unmanned autonomous helicopter system are considered in the design of sensor fault-tolerant control. The inequality constraint that the output tracking error must satisfy is transformed into an unconstrained problem by introducing an error transformation function. The fault state of the velocity sensor is considered as the system uncertainty, and a neural network is used to approach the total uncertainty. Neural network estimation errors and external disturbances are treated as compound disturbances, and a high-order disturbance observer is constructed to compensate for them.

Agriculture is the backbone of a country, contributing more than half of the sector of economy throughout the world. The need for precision agriculture is essential in evaluating the conditions of the crops with the aim of determining the proper selection of pesticides. The conventional method of pest detection fails to be stable and provides limited accuracy in the prediction. This paper aims to propose an automatic pest detection module for the accurate detection of pests using the hybrid optimization controlled deep learning model.

The paper proposes an advanced pest detection strategy based on deep learning strategy through wireless sensor network (WSN) in the agricultural fields. Initially, the WSN consisting of number of nodes and a sink are clustered as number of clusters. Each cluster comprises a cluster head (CH) and a number of nodes, where the CH involves in the transfer of data to the sink node of the WSN and the CH is selected using the fractional ant bee colony optimization (FABC) algorithm. The routing process is executed using the protruder optimization algorithm that helps in the transfer of image data to the sink node through the optimal CH. The sink node acts as the data aggregator and the collection of image data thus obtained acts as the input database to be processed to find the type of pest in the agricultural field. The image data is pre-processed to remove the artifacts present in the image and the pre-processed image is then subjected to feature extraction process, through which the significant local directional pattern, local binary pattern, local optimal-oriented pattern (LOOP) and local ternary pattern (LTP) features are extracted. The extracted features are then fed to the deep-convolutional neural network (CNN) in such a way to detect the type of pests in the agricultural field. The weights of the deep-CNN are tuned optimally using the proposed MFGHO optimization algorithm that is developed with the combined characteristics of navigating search agents and the swarming search agents.

The analysis using insect identification from habitus image Database based on the performance metrics, such as accuracy, specificity and sensitivity, reveals the effectiveness of the proposed MFGHO-based deep-CNN in detecting the pests in crops. The analysis proves that the proposed classifier using the FABC+protruder optimization-based data aggregation strategy obtains an accuracy of 94.3482%, sensitivity of 93.3247% and the specificity of 94.5263%, which is high as compared to the existing methods.

The proposed MFGHO optimization-based deep-CNN is used for the detection of pest in the crop fields to ensure the better selection of proper cost-effective pesticides for the crop fields in such a way to increase the production. The proposed MFGHO algorithm is developed with the integrated characteristic features of navigating search agents and the swarming search agents in such a way to facilitate the optimal tuning of the hyperparameters in the deep-CNN classifier for the detection of pests in the crop fields.

The purpose of this paper is to compare the influence of relative density (RD) and strain rates on failure mechanism and specific energy absorption (SEA) of polyamide lattices ranging from bending to stretch-dominated structures using selective laser sintering (SLS).

Three bending and two stretch-dominated unit cells were selected based on the Maxwell stability criterion. Lattices were designed with three RD and fabricated by SLS technique using PA12 material. Quasi-static compression tests with three strain rates were carried out using Taguchi's L9 experiments. The lattice compressive behaviour was verified with the Gibson–Ashby analytical model.

It has been observed that RD and strain rates played a vital role in lattice compressive properties by controlling failure mechanisms, resulting in distinct post-yielding responses as fluctuating and stable hardening in the plateau region. Analysis of variance (ANOVA) displayed the significant impact of RD and emphasised dissimilar influences of strain rate that vary to cell topology. Bending-dominated lattices showed better compressive properties than stretch-dominated lattices. The interesting observation is that stretch-dominated lattices with over-stiff topology exhibited less compressive properties contrary to the Maxwell stability criterion, whereas strain rate has less influence on the SEA of face-centered and body-centered cubic unit cells with vertical and horizontal struts (FBCCXYZ).

This comparative study is expected to provide new prospects for designing end-user parts that undergo various impact conditions like automotive bumpers and evolving techniques like hybrid and functionally graded lattices.

To the best of the authors' knowledge, this is the first work that relates the strain rate with compressive properties and also highlights the lattice behaviour transformation from ductile to brittle while the increase of RD and strain rate analytically using the Gibson–Ashby analytical model.

This study aims to learn the dynamic radar cross-section (RCS) of a deflection air brake.

The aircraft model with delta wing, V-shaped tail and blended wing body is designed, and high-precision unstructured grid technology is used to deal with the surface of air brake and fuselage. The calculation method based on multiple tracking and dynamic scattering is presented to calculate RCS.

The fuselage has a low scattering level, and the opening air brake will bring obvious dynamic RCS effects to itself and the whole machine. The average indicator of air brake RCS can be lower than –0.6 dBm² under the tail azimuth, while that of forward and lateral direction is lower. The mean RCS of fuselage is obviously higher than that of air brake, while the deflected air brake and its cabin can still provide strong scattering sources at some azimuths. When the air brake is opening, the change amplitude of the aircraft forward RCS can exceed 19.81 dBm².

This research has practical significance for the dynamic electromagnetic scattering analysis and stealth design of the air brake.

The calculation method for aircraft RCS considering air brake dynamic deflection has been established.

Digitalisation, which involves the use of digital technologies in transforming an organisation’s activities, transcends just the acquiring of emerging digital tools. Having the right people to drive the implementation of these technologies and attaining strategic organisational goals is essential. While most studies have focused on the use of emerging technologies in the construction industry, less attention has been given to the ‘people’ dimension. Therefore, this study aims to assess the people-related features needed for construction digitalisation.

The study adopted pragmatic thinking using a mixed-method approach. A Delphi was used to achieve the qualitative aspect of the research, while a questionnaire survey conducted among 222 construction professionals was used to achieve the quantitative aspect. The data gathered were analysed using frequency, percentage, mean item score, Kruskal–Wallis H test, exploratory factor analysis and confirmatory factor analysis.

Based on acceptable reliability, validity and model fit indices, the study found that the people-related factors needed for construction digitalisation can be grouped into technical capability of personnel, attracting and retaining digital talent and organisation’s digital culture.

The findings offer valuable benefits to construction organisations as understanding these identified people features can help lead to better deployment of digital tools and the attainment of the digital transformation.

This study attempts to fill the gap in the shortage of literature exploring the people dimension of construction digitalisation. The study offers an excellent theoretical backdrop for future works on digital talent for construction digitalisation, which has gained less attention in the current construction digitalisation discourse.

Rebar is the prime component of reinforced concrete structures, and rebar monitoring is a time-consuming and technical job. With the emergence of the fourth industrial revolution, the construction industry practices have evolved toward digitalization. Still, hesitation remains among stakeholders toward the adoption of advanced technologies and one of the significant reasons is the unavailability of knowledge frameworks and implementation guidelines. This study aims to investigate technical factors impacting automated monitoring of rebar for the understanding, confidence gain and effective implementation by construction industry stakeholders.

A structured study pipeline has been adopted, which includes a systematic literature collection, semistructured interviews, pilot survey, questionnaire survey and statistical analyses via merging two techniques, i.e. structural equation modeling and relative importance index.

The achieved model highlights “digital images” and “scanning” as two main categories being adopted for automated rebar monitoring. Moreover, “external influence”, “data-capturing”, “image quality”, and “environment” have been identified as the main factors under “digital images”. On the other hand, “object distance”, “rebar shape”, “occlusion” and “rebar spacing” have been highlighted as the main contributing factors under “scanning”.

The study provides a base guideline for the construction industry stakeholders to gain confidence in automated monitoring of rebar via vision-based technologies and effective implementation of the progress-monitoring processes. This study, via structured data collection, performed qualitative and quantitative analyses to investigate technical factors for effective rebar monitoring via vision-based technologies in the form of a mathematical model.