Computer Engineering Commons^™

A Novel Reinforcement Learning Method For Efficient Cross-Training Between Real And Simulated Robots, Zhonglin Liang

Master's Theses

Training reinforcement learning policy in a simulated environment can be a go-to choice for many research topics, as a simulated environment provides flexibility and costs less than building a physical environment for training. However, policy trained in a simulated environment often fails to transfer to the real environment for problems that have a more complex dynamics. As the solution, Sim2Real was proposed and it categorizes techniques that improve the performance of the transfer from simulation to reality. Sim2Real is an area of study that focuses on utilizing simulation data to train models that can apply to real world environment. It …

Investigating Spatiotemporal Trends Using Precursory Signatures: Implications To Develop Short-Term Earthquake Forecasting Techniques In Sumatra-Andaman Region, Ramya Jeyaraman J

Theses and Dissertations

Earthquake forecasting is a challenging field due to Earth's heterogeneous nature. This research aims to develop a short-term earthquake forecasting model by analyzing spatiotemporal trends and precursory signatures in the Sumatra-Andaman region, known for its high seismic activity and tsunami risk. The study adopts an interdisciplinary approach, integrating solid earth tides (SET), micro shocks, and outgoing longwave radiation (OLR) to gain deeper insights into seismic nucleation processes. The research begins by using Singular Spectral Analysis (SSA) to identify potential seismically vulnerable areas through the analysis of irregularities in SET.

A spatiotemporal analysis of micro shocks is conducted to assess the …

Enhancing Cross-Cultural Communication In Low-Resource Language Conversational Agents, Hardi V. Trivedi

Master's Theses

Recent advancements in natural language processing (NLP) and large language models (LLMs) have facilitated the development of systems capable of generating human-like responses across a wide range of tasks. However, the majority of research has focused predominantly on English, overlooking the vast linguistic diversity globally. For true global inclusivity, extending research to other languages is crucial, particularly as it can significantly benefit various sectors such as business, healthcare, government, and education. A major challenge in this expansion is the scarcity of digital data available and the limited number of pre-trained models for low-resource languages. Our research specifically addresses these challenges …

Sustainable Mobility: Machine Learning-Driven Deployment Of Ev Charging Points In Dublin, Ruairí De Fréin, Alexander Mutiso Mutua Mr

Articles

Electric vehicle (EV) drivers in urban areas face range anxiety due to the fear of running out of charge without timely access to charging points (CPs). The lack of sufficient numbers of CPs has hindered EV adoption and negatively impacted the progress of sustainable mobility. We propose a CP distribution algorithm that is machine learning-based and leverages population density, points of interest (POIs), and the most used roads as input parameters to determine the best locations for deploying CPs. The objects of the following research are as follows: (1) to allocate weights to the three parameters in a $6$ km …

An Open Approach To Clinical Study Recruitment Automation, Andrew Bouras, Andrew Bouras, Dhruv Patel, Rahul Kataria

HCA-NSU MD Research Day

Objective: To present an innovative Patient Recruitment Automation System that leverages open standards and promotes open data in clinical research. Background: Patient recruitment for clinical studies faces challenges in efficiency, data security, and regulatory compliance. These issues can delay research timelines and impact healthcare advancements. Methods: We developed a system that integrates with Electronic Health Record (EHR) systems to identify potential participants using predefined criteria. The platform employs Azure-based infrastructure with advanced data protection techniques, ensuring HIPAA compliance. It features an automated screening process using adaptive questionnaires and a communication module for participant engagement. Results: The system demonstrates improved efficiency …

Collapse Of Pre-Covid-19 Differences In Performance In Online Vs. In-Person College Science Classes, And Continued Decline In Student Learning, Gregg R. Davidson, Hong Xiao, Kristin Davidson

Faculty and Student Publications

Abstract: Studies comparing student outcomes for online vs. in-person classes have reported mixed results, though with a majority finding that lower-performing students, on average, fare worse in online classes, attributed to the lack of built-in structure provided by in-person instruction. The online/in-person outcome disparity was normative for non-major geology classes at the University of Mississippi prior to COVID-19, but the difference disappeared in the years after 2020. Previously distinct trendlines of GPA-based predictions of earned-grade for online and in-person classes merged. Of particular concern, outcomes for in-person classes declined to match pre-COVID-19 online expectations, with lower-GPA students disproportionally impacted. Objective …

A Method For Key Node Identification In Operational Target System Based On War Gaming, Yongfu Zhang, Yang Liu, He Yuan

Journal of System Simulation

Abstract: The identification of key nodes in an operational target system is an important basis for combat command decision-making. Due to the lack of experimental verification of key node identification in the current operational target system in a campaign-level dynamic confrontation environment, a complex network model of operational target system with large-scale entities and complex interaction relationship was constructed by taking integrated air defense network as an example, with the help of the data derived from the large joint war gaming; the characteristics of wargame data were considered, and the value characteristics of combat targets and network structure characteristics were …

Platform Path Optimization Method Based On Cumulative Detection Probability Of Sonar Search, Xiang Wei, Xingxuan Liu, Dianzheng Fu, Tianji Yang, Jiaxuan Yang

Journal of System Simulation

Abstract: To address the lack of research on the optimal path of mobile search platform to search for moving targets, this paper proposes a path optimization method of mobile search platform based on cumulative search probability theory. Based on the cumulative detection probability (CDP), one of the important criteria of sensor performance evaluation, a single-peak CDP calculation formula is constructed by using a time series correlation model, namely the (λ, σ) process model. A set of target motion scenarios are constructed, and the trajectory probability of target scenarios and their CDP at different time are corrected by Bayesian posterior probability. …

Modeling For Decision Support Of Flight Ground Support Process, Zhiwei Xing, Ruiwen Yu, Biao Li, Zhaoxin Chen

Journal of System Simulation

Abstract: Aiming at the problems of insufficient decision-making ability and low operational efficiency of the flight ground support process, a decision support model of the flight ground support process based on the department of defense architecture framework (DoDAF) is proposed. Starting from the support operation, support resources, and the relationship between them, the quantitative description of the flight ground support process is performed. DoDAF and the model-based systems engineering (MBSE) modeling method are combined to establish a decision support model of the flight ground support process. The decision utility function is established to analyze the utility value of the comprehensive …

End-To-End Motion Planning Of Unmanned Vehicles Based On Multimodal Deep Reinforcement Learning, Kaiyuan Ding, Askar Hamdulla, Bin Zhu, Eksan Firkat, Zhengtang Ma

Journal of System Simulation

Abstract: Since the agent cannot sense the surrounding environment and cannot successfully avoid obstacles, reinforcement learning fails to be generalized to robot motion planning in difficult terrain. Therefore, a solution based on multimodal deep reinforcement learning, which learns to blend proprioceptive states with high-dimensional depth sensor inputs, is proposed for the motion planning of unmanned vehicles. To be specific, proprioceptive states offer contact measurement for immediate reaction, and the unmanned vehicle can learn and forecast environmental changes with its attached visual sensors, proactively navigating around obstacles and uneven terrains numerous time steps ahead. TransProAct (transformer-based proactive action), a unique end-to-end …

Flexible Job Shop Scheduling Method Based On Collaborative Agent Reinforcement Learning Algorithm, Jian Li, Huankun Li, Pengbo He, Huabei Wang, Liping Xu, Kui He

Journal of System Simulation

Abstract: To enhance the efficiency of flexible job shop scheduling, this paper develops a Markov decision process with specific constraints tailored to the scheduling problem. A cooperative agent reinforcement learning method is proposed to solve the problem of concurrent selection of workpieces and machines. During the construction of the Markov decision process, a disjunctive graph is introduced to represent the state characteristics. Two agents are introduced to select the workpieces and machines. The reward parameters governing the entire scheduling process are established by predicting variations in the minimum-maximum completion time across different time points. A GIN(graph isomorphic network) graph neural …

Dual-Resource Constrained Distributed Flexible Scheduling For Aerospace Structural Components, Yufang Wang, Dianqing Zhang, Xiaolin Hua, Binbin Yao, Fan Chen

Journal of System Simulation

Abstract: A dual-resource constrained distributed flexible job-shop scheduling model was established by taking into account the worker constraints of the finishing process and the requirements of distributed multi-factory collaboration in the production of aerospace structural components. A hybrid grey wolf optimization algorithm based on the critical factory was proposed to solve this problem. The model contained four subproblems: factory selection, operation sequencing, machine selection, and worker selection. In view of these four sub-problems, a four-layer coding and a new decoding method were designed to avoid the use conflict of machines and workers. In addition, a new mechanism for hunting and …

Automatic Detection Algorithm For Typical Defects Of Substation Based On Improved Yolov5, Zhongkai Xu, Yanling Liu, Xiaojuan Sheng, Chao Wang, Wenjun Ke

Journal of System Simulation

Abstract: In response to the challenges present in the context of defect recognition in substations, such as complex substation defects and sample imbalance, an improved YOLOv5 algorithm was proposed. The Transformer model was introduced into the YOLOv5 network structure, leveraging the self-attention mechanism to capture long-range dependencies among features. A focal loss-based optimization was employed to improve the loss function, as well as the detection accuracy and robustness of defects of small sample substations. To meet the requirements of substation defect recognition, a dedicated dataset was constructed. A clustering algorithm was applied to the real annotation boxes to generate more …

Object Detection Of Lightweight Transformer Based On Knowledge Distillation, Gaihua Wang, Kehong Li, Qian Long, Jingxuan Yao, Bolun Zhu, Zhengshu Zhou, Xuran Pan

Journal of System Simulation

Abstract: In autonomous driving, the efficiency and accuracy of object detection are significant. Object detection based on Transformer structure has gradually become the mainstream method, eliminating the complex anchor generation and non-maximum suppression (NMS). It has problems of high computing cost and slow convergence. An object detection model of the based lightweight pooling transformer (LPT) is designed, which contains a pooling backbone network and dual pooling attention mechanism. A general knowledge distillation method is intended for the DETR (detection transformer) model, which transfers prediction results, query vector, and features extracted by the teacher as knowledge to the LPT model to …

A Hybrid Genetic Search Algorithm For Capacitated Electric Vehicle Routing Problem, Dongyao Jin, Mi Liu, Yena Zhu, Yijiang Zhao

Journal of System Simulation

Abstract: The capacitated electric vehicle routing problem (CEVRP) is an NP-hard combinatorial optimization problem in logistics distribution, aiming to minimize the total delivery distance of electric vehicles while satisfying carrying capacity and battery charge constraints. A hybrid genetic search algorithm is proposed to solve CEVRP by decomposing it into two subproblems: capacitated vehicle routing problem (CVRP) and fixed-route vehicle charging problem (FRVCP). A coding scheme with a two-layer chromosome structure is designed to represent the decision variables of these two subproblems. A Split operation is employed to generate vehicle routes for solving CVRP, and five neighborhood search operators, including Relocate, …

Traffic Sign Recognition Model With Long-Tail Distribution Based On Yolox-Tiny, Yunpeng Wu, Yingxiong Fu, Lijun Shen, Feng Cui

Journal of System Simulation

Abstract: Accurate recognition of traffic signs plays an important role in the field of intelligent driving. Traffic sign training datasets with long-tail distribution increase the difficulty of traffic sign recognition. A traffic sign recognition model with long-tail distribution based on YOLOX-Tiny was proposed to improve the poor performance of the model trained on long-tail distribution datasets. A long-tail traffic sign dataset was created based on the TT100K_2021 (tsinghua-tencent 100K 2021) dataset. YOLOX-Tiny was chosen as the underlying model by considering picture numbers in datasets, sample distribution, and model size. Equalization loss v2 (EQL v2) was used as classification loss to …

Real-Time Lidar Slam Algorithm Based On Distribution Optimal Registration, Weigang Li, Chuxiang Yu, Yongqiang Wang, Shaofeng Zou

Journal of System Simulation

Abstract: When scanning the surrounding environment, a lidar will generate some cluttered and sparse point cloud, which will cause excessive distribution fitting errors and correlation distances in the registration process, thus affecting the accuracy of the registration algorithm and the effect of simultaneous localization and mapping (SLAM). To address this problem, a real-time lidar SLAM algorithm based on distribution optimal registration is proposed. An eigenspectrum filter is designed, which takes the normalized minimum eigenvalue as the filtering object to filter out the points that do not match the set distribution in order to reduce the distribution fitting error. Secondly, a …

Research On Green Job Shop Scheduling Based On Herd Immunity Optimizer, Xunde Ma, Li Bi, Junjie Wang

Journal of System Simulation

Abstract: In view of the green flexible job shop scheduling problem where machines have multiple speeds, a green flexible job shop scheduling model under multiple speeds was constructed to minimize the makespan and total energy consumption under different speeds. A discrete coronavirus herd immunity optimizer (DCHIO) was proposed for a solution. A discrete individual updating method was introduced for the relatively large solution space of the multi-speed problem, based on which a population updating mechanism with multi-scale joint search was proposed to search the solution space quickly and uniformly. A dynamic mutation operation was designed to enhance the population diversity …

Agv Scheduling Problem At Automated Terminals Based On Improved Dqn Algorithm, Chengji Liang, Shidong Zhang, Yu Wang, Bin Lu

Journal of System Simulation

Abstract: A future tasks considering deep Q-network (F-DQN) algorithm was proposed to output realtime scheduling results of automated guided vehicles (AGVs) at automated terminals. This algorithm combined the advantages of real-time scheduling and static scheduling, improving the system status by considering static future task information when making real-time decisions, so as to obtain a better scheduling solution. In this study, the actual layout and equipment conditions of the Yangshan phase IV automated terminal were considered, and a series of simulation experiments were conducted using the Plant Simulation software. The experimental results show that the F-DQN algorithm can effectively solve the …

Global-Local Fusion For Efficient 3d Object Detection, Bin Lu, Minghan Wang, Yang Sun, Zhenyu Yang

Journal of System Simulation

Abstract: As the 3D object detection based on point clouds shows an incapacity of feature extraction and incongruity between classification and regression, this research introduces a novel ResCST architecture based on the SECOND network. It incorporates residual connections into the 3D sparse convolutional layer, with the advantages of capturing long-distance dependent relation by SwinTransformer and obtaining local features by convolutional neural network integrated, proposing the CNN-SwinTransformer hybrid model for enhanced feature extraction. It introduces the RCIoU method for the joint optimization of classification and regression tasks. The experimental results show that the model achieves a 3D detection accuracy of 91.21%, …

Optimization Of Crucial Targets For Air Defense Based On Combined Weighting-Topsis Model, Peng Zhang, Ke Feng

Journal of System Simulation

Abstract: To optimize the selection of crucial defended targets in regional air defense operations and improve the selection accuracy, this paper constructs a factor model considering the importance of air defense targets from the perspectives of target value, defense urgency, target vulnerability, and target recovery. Under the optimization of the TOPSIS method through a combination of the ANP and entropy weight methods, tendentious opinions of commanders and the excessive reliance on objective data can be overcome to ensure the factor weighting is more reasonable and accurate; Super Decisions is used to calculate the weights of the ANP method, accelerating data …

Eecbs Multi-Robot Path Planning Based On Variable Suboptimal Factors Of Prioritizing Conflicts, Xingyu Yan, Niya Wang, Jianlin Mao, Zhigang He, Dayan Li

Journal of System Simulation

Abstract: In the process of multi-robot path planning (MRPP), the unavoidable key conflicts between the optimal paths have a significant impact on the efficiency of path solving. To address this issue, an MRPP method based on variable suboptimal factors of prioritizing conflicts (PC) was proposed. Path search was performed at the lower layer of the explicit estimation conflict-based search (EECBS) algorithm; in the upper layer of the EECBS algorithm framework, the PC was determined, and the suboptimal factor of the robot with key conflicts was adaptively increased; by analyzing the distribution of obstacles in the surrounding neighborhoods of key conflicts, …

Slam Dynamic Algorithm Based On Improved Feature Description, Qiang Fu, Xianyun Teng, Yuanfa Ji, Fenghua Ren

Journal of System Simulation

Abstract: The original ORB descriptor algorithm has a low matching accuracy and long matching time, the positioning accuracy and robustness of the SLAM(simultaneous localization and mapping) system are severely disturbed by moving objects in dynamic scenes, and the ORB-SLAM3 system is incapable of constructing dense maps. To address the above problems, this paper proposes an improved ORB-SLAM3 based on the BEBLID descriptor and object detection. A lightweight YOLOv5s dynamic object detection network and dynamic feature removal module are fused with the tracking thread to improve the system's positioning accuracy. Replacing the original feature description algorithm, an improved local image descriptor …

Design And Implementation Of Maritime Unmanned Cross-Domain Collaborative Effectiveness Evaluation System Based On Mlp, Hongyu Hu, Tianzhu Gao, Haitao Gu

Journal of System Simulation

Abstract: In the face of the problem of evaluating the detection capability effectiveness of the maritime unmanned cross-domain collaborative system, it is necessary to study the evaluation indexes and evaluation algorithm. In this paper, the robot's own parameters and environmental parameters are combined to build a calculation model for evaluation indexes, such as detection coverage rate, repeated detection rate, the number of pixels per unit area, and energy as well as an evaluation system for detection capability of the maritime unmanned cross-domain collaborative system. The subjectivity in the evaluation process is reduced, and training data is generated by the availability …

Improvement Of A* Algorithm In Path Planning Of Mobile Robot, Dexin Yao, Hongjun San, Yaru Wang, Haijie Sun, Jiupeng Chen, Xiaoyuan Yang

Journal of System Simulation

Abstract: To solve the problems of excessively redundant nodes, low search efficiency, and excessive path turning angle in the path search of the traditional A* algorithm, an improved A* algorithm is proposed to plan the optimal path. First, the amount of search neighborhood of the A* algorithm is increased to 24 to obtain a more accurate and comprehensive search field. Second, the angle search algorithm is introduced, eliminating the unnecessary nodes in the path search and making the search more target-oriented. Third, the heuristic function is weighted by the exponential attenuation through the relative position of the current point and …

Modeling And Simulation Of The 12-Meter Antenna, Antonio Pires

Morehead State Theses and Dissertations

A thesis presented to the faculty of the College of Science at Morehead State University in partial fulfillment of the requirements for the Degree of Master of Science by Antonio Pires on November 11, 2024.

Training An Ai To Detect Injection Attacks Using A Hands On Approach, Aedan Tredinnick

Cybersecurity Undergraduate Research Showcase

This paper presents a practical approach to training an AI model to detect injection attacks, focusing on the creation of a manufactured dataset via structured hands-on methods. By establishing a vulnerable web server using XAMPP and DVWA (Damn Vulnerable Web Application), the research aims to simulate various injection attacks and capture relevant network traffic data. The paper discusses the methodology of data collection, AI model development, and performance evaluation.

Importance Of Soft Skills Comparative Study With Cybersecurity Professionals In The Manufacturing And Finance Critical Sectors, Stanley Mierzwa, Mary Lind

Center for Cybersecurity

Cybersecurity professionals require and will benefit from having strong and competent technical and soft skills, knowledge, and abilities. Cyber-attacks continue to plague our organizations and businesses, and finding the individuals with the skills needed for industry teams to contend with these broad and varying types of breaches is essential. This research article will outline the results of a comparative quantitative study that compares the importance of soft skills or nontechnical competencies by information security and cybersecurity professionals in the finance and manufacturing critical infrastructure sectors. This research study used a validated survey instrument from a previous seminal study to capture …

A Benchmark Knowledge Graph Of Driving Scenes For Knowledge Completion Tasks, Ruwan Wickramarachchi, Cory Henson, Amit Sheth

Publications

Knowledge graph completion (KGC) is a problem of significant importance due to the inherent incompleteness in knowledge graphs (KGs). The current approaches for KGC using link prediction (LP) mostly rely on a common set of benchmark datasets that are quite different from real-world industrial KGs. Therefore, the adaptability of current LP methods for real-world KGs and domain-specific ap- plications is questionable. To support the evaluation of current and future LP and KGC methods for industrial KGs, we introduce DSceneKG, a suite of real-world driving scene knowledge graphs that are currently being used across various industrial applications. The DSceneKG is publicly …

If Androids Dream, Are They More Than Sheep?: Westworld, Robots And Legal Rights, Amanda J. Dipaolo

Dialogue: The Interdisciplinary Journal of Popular Culture and Pedagogy

The robot protagonists in HBO’s Westworld open the door to several philosophical and ethical questions, perhaps the most complex being: should androids be granted similar legal protections as people? Westworld offers its own exploration of what it means to be a person and places emphasis on one’s ability to feel and understand pain. With scientists and corporations actively working toward a future that includes robots that can display emotion in a way that can convincingly pass as that of a person’s, what happens when androids pass the Turing test, feel empathy, gain consciousness, are sentient, or develop free will? The …