Digital Thread: Bridging Macro–Micro Services In System-Of-Systems,
2025
Old Dominion University
Digital Thread: Bridging Macro–Micro Services In System-Of-Systems, Joshit Mohanty
Graduate Student Government Association Research Conference
Organizations and industries increasingly rely on distributed services in decentralized environments—ranging from large-scale, system-of-system architectures to fine-grained, agent-based microservices. While this distributed paradigm offers flexibility and innovation, it presents critical challenges such as interoperability gaps, inconsistent data formats, and a lack of holistic oversight. Traditional integration approaches, including ad-hoc middleware or enterprise service buses, tend to solve these issues reactively. As a result, technical debt accumulates, stakeholder misalignments persist, and scaling to new demands becomes complex.
This research proposes digital thread (DT) as the unifying framework to create an authoritative source of truth: a continuous flow of information across the …
Meta-Clustering For Specialized Language Models: Enhancing Contextual Adaptation And Mitigating Hallucinations In Diverse Healthcare Environments,
2025
Old Dominion University
Meta-Clustering For Specialized Language Models: Enhancing Contextual Adaptation And Mitigating Hallucinations In Diverse Healthcare Environments, Joshit Mohanty, Vaishali Vaishali, Sandeep Kumar Nayak, Sumit Lahiri
Graduate Student Government Association Research Conference
Large Language Models (LLMs) have significantly advanced conversational AI by enabling dialogic information-seeking and task execution across diverse domains. However, their extensive parameters and broad domain scope lead to “data hallucinations.” These shortcomings are particularly evident in dynamic and diverse environments like India’s healthcare sector, where myriad languages, regional practices, and cultural nuances demand specialized, localized expertise rather than one-size-fits-all generalist models. This paper introduces a meta-clustering framework that integrates Distilled Language Models (DLMs) and Small/Specialized Language Models (SLMs) with meta-learning principles to address these limitations. By drawing on evidence from works such as MedHalu and Med-HALT, the framework seeks …
Tamos: Task-Aware Multi-Agent Orchestrator System,
2025
Indian Institute of Technology Kanpur
Tamos: Task-Aware Multi-Agent Orchestrator System, Joshit Mohanty, Sandeep Kumar Nayak, Sumit Lahiri
Graduate Student Government Association Research Conference
Large language models (LLMs) are increasingly at the core of multi-agent systems (MAS). However, the high resource demand, error propagation, and lack of adaptive evaluation mechanisms pose significant challenges in deploying these agentic solutions at scale. To address these concerns, this research proposes a Task-Aware Multi-Agent Orchestrator System designed to refine the agentic framework, categorizing tasks autonomously, assigning specialized evaluation datasets, and balancing token usage against functional effectiveness. This approach underscores robust data management, including AsyncHow, Mosaic AI, and Synthetic Preference Optimization (PO) corpora. Each dataset targets specific dimensions of agent performance, such as dynamic task decomposition and tool integration …
Graphtreemed: A Hybrid Graph-Tree Rag Architecture For Mission-Critical Medical Applications,
2025
Indian Institute of Technology Kanpur
Graphtreemed: A Hybrid Graph-Tree Rag Architecture For Mission-Critical Medical Applications, Joshit Mohanty, Sandeep Kumar Nayak, Sumit Lahiri
Graduate Student Government Association Research Conference
Studies within engineering management indicate that decision-making is often based on the cognitive processing of grouped and pictographic information clusters entangled with high-level pattern recognition. Similarly, graph-based retrieval-augmented generation (RAG) architectures substantially improve diagnostic accuracy and interpretability, while tree-structured systems reduce critical misses through hierarchical reasoning. However, existing solutions often lack a unified framework that seamlessly integrates these two paradigms to address the multifaceted demands of mission-critical healthcare settings. This proposal introduces GraphTreeMed, a novel hybrid RAG architecture designed to harness the complementary strengths of graph-based and tree-based retrieval mechanisms, thereby advancing the safety and efficacy of clinical decision support …
Perceptions Of Employability With Ai Skills,
2025
Lynn University
Perceptions Of Employability With Ai Skills, Brandy Whitford, Patrick J. Cooper
Student Publications and Presentations
Artificial intelligence (AI) is making AI proficiency a key factor in hiring and career advancement. By late 2023, 75% of knowledge workers integrated AI into their workflows, with 92% reporting increased productivity and creativity (Kimbrough, 2024). Employers are adapting—66% prefer candidates with AI expertise, and 77% consider AI skills essential for career growth (Microsoft & LinkedIn, 2024). However, hiring biases related to AI-skilled applicants remain underexplored, particularly concerning gender disparities in employability perceptions. This study examines how AI-related skills influence perceived employability and whether these perceptions vary based on applicant gender. Specifically, it explores whether AI-skilled female applicants receive higher …
Gender Bias Within Ai Imaging,
2025
Lynn University
Gender Bias Within Ai Imaging, Drew Quattrocchi
Student Publications and Presentations
This study investigates AI-created gender bias in AI-created images through content analysis, contrasting the way gender is depicted in professions in leading AI image-creation tools such as Chat smith, Adobe Firefly, Midjourney, and Stable Diffusion. Employing a quantitative research method, this study contrasts AI-created images of gender-stereotypical careers for both male and female. Non-gendered careers will be used as well to identify patterns of stereotyping and bias. The area of emphasis lies in individual subjects within the images and scrutinizing visual elements such as clothing, accessories, background, face expressions, and gendered roles assigned to each. Particular emphasis is focused to …
Survey On Intelligent Planning Methods From Large Language Models Perspective,
2025
College of Intelligence Science and Technology, National University of Defense Technology, Changsha 410073, China
Survey On Intelligent Planning Methods From Large Language Models Perspective, Yanzhong Zhou, Junren Luo, Xueqiang Gu, Wanpeng Zhang
Journal of System Simulation
Abstract: Starting from the perspective of large language models, this paper gives an overview of the definition and development of intelligent planning, and briefly introduces the traditional methods of intelligent planning; based on the close relationship between large language model intelligent agents and intelligent planning, introduces the architecture of large language models and typical large model intelligent agents; focusing on the intelligent planning for large language models, combs through the learning of planning languages, chain of thought, feedback optimization, and process automation; combining with the current challenges and difficulties, introduces the outlook of cutting-edge research on intelligent planning with large …
Simulation Environment Construction Of Track Segment Association And Algorithm Performance Evaluation,
2025
School of Information and Electronic Engineering, Shandong Technology and Business University, Yantai 264005, China
Simulation Environment Construction Of Track Segment Association And Algorithm Performance Evaluation, Dian Ding, Guangfen Wei, Zheng Cao, Shaohui Wen
Journal of System Simulation
Abstract: In order to study the applicability of Track Segment Association (TSA) algorithms in actual radar working environment , a TSA simulation environment which can simulate the real movement of the target is constructed. By constructing a rich set of target motion sets, the state switching process of target motion is described based on Markov state transition matrix, and the density is flexibly controlled through track translation. The simulation results show that this environment can evaluate the performance of the current classical TSA algorithms. The evaluation results provide a good reference for the practical engineering application of interrupted track association.
Research On The Resilience Of Integrated Urban Passenger Transport Network In Urban Agglomerations Considering The Intra-Urban Service Network,
2025
School of Transportation, Inner Mongolia University, Hohhot 010070, China
Research On The Resilience Of Integrated Urban Passenger Transport Network In Urban Agglomerations Considering The Intra-Urban Service Network, Shida Nie, Chengbing Li, Bowei He, Xintao Li
Journal of System Simulation
Abstract: In order to solve the problem of insufficient comprehensiveness and refinement of the urban agglomeration passenger transport network model, Space L modelling method in complex network theory is adopted to construct a comprehensive urban passenger transport network model considering the urban internal service network. A comprehensive urban passenger transport network composed of intercity networks and intra-city service networks is built, and time-dependent edge weights in the network is considered. A time-weighted network efficiency model is proposed to evaluate the network resilience. The results show that under a random attack strategy, the relative time efficiency of the network fluctuates less, …
A Multi-Robot Collaborative Path Planning Algorithm With Chain Working Mode,
2025
Faculty of Information Engineering and Automation, Kunming University of Science and Technology, Kunming 650500, China
A Multi-Robot Collaborative Path Planning Algorithm With Chain Working Mode, Zhigang He, Dayan Li, Niya Wang, Jianlin Mao, Ning Wang
Journal of System Simulation
Abstract: In order to solve the problem that the traditional MAPF algorithms can lead to a large number of repeated paths and thus non-essential energy loss in application scenarios where multiple robots have a common goal point, a multi-robot chain work mode with a tractor is proposed, which divides the robots with common target points into subgroups for multi robot collaborative path planning, and a collaborative dynamic priority SIPP with tractor (Co-DPtSIPP) algorithm is given. The polygonal Fermat point principle and other methods are used to obtain the serial connection areas of each collaborative group robot; considering the sequence of …
Traffic Signal Detection Based On Improved Yolov7,
2025
School of Mathematics and Physics, China University of Geosciences, Wuhan 430074, China
Traffic Signal Detection Based On Improved Yolov7, Lanyue Zheng, Yujie Zhang
Journal of System Simulation
Abstract: An improved YOLOv7 is proposed to address the problem of low recognition accuracy in general object detection algorithms for traffic signal detection. The algorithm removes the 20×20 detection scale and adds a 160×160 detection scale to increase shallow features while making the model lightweight. It combines the bi-level routing attention (BRA) proposed in BiFormer with axial attention, and innovatively proposes axially-guided BRA (ABRA). This module is specifically designed for the characteristics of traffic signal positions. To address the issue of object size sensitivity to the IoU metric, the normalized wasserstein distance (NWD) measurement is introduced to improve object location …
Research On Dual-Layer Path Planning Method For Lunar Rover Based On Slip Prediction,
2025
School of Mathematics, China University of Mining and Technology, Xuzhou 221116, China
Research On Dual-Layer Path Planning Method For Lunar Rover Based On Slip Prediction, Zhang Xingyu, Baolei Wu, Jun Wang, Miaoying Hong, Jiahui Wang, Yongqiang Qi
Journal of System Simulation
Abstract: In response to the challenges faced by lunar rovers in the process of path planning, such as safe obstacle avoidance and target deviation caused by complex terrain, a dual-layer path planning based on slip prediction is proposed. In this approach, flat terrain is adaptively selected to reduce the wheel slip of the lunar rover. The overall complexity of the terrain is calculated using digital elevation information, and a Q-learning algorithm with a three-level reward mechanism is designed to navigate around highslip areas, achieving global path planning. A depth camera is used to perceive obstacles, a dynamic window method based …
Design Of Distributed Multi-Functional Integrated Signal-Level Confrontation Simulation System In Local Area,
2025
Nanjing Research Institute of Electronics Technology, Nanjing 210039, China
Design Of Distributed Multi-Functional Integrated Signal-Level Confrontation Simulation System In Local Area, Weiqian Li, Tianyu Yang, Zongyang Li, Jianjun Chen
Journal of System Simulation
Abstract: In order to study the resources management and self-organized collaborative application method of multiple multi-functional integrated electronic equipment in the region, we build a signal-level digital simulation system that supports multiple distributed multi-functional integrated electronic equipment within a region to carry out cooperation or confrontation. A joint time advancing mechanism named "variable-step time advancing method based on frame scheduling" and "independent event driven time advancing method " is proposed. It can not only ensure the integrity of each frame of radar simulation data for each equipment, but also enable multiple equipment in the simulation system to advance simultaneously and …
Uav Path Planning Based On Improved Deep Deterministic Policy Gradients,
2025
College of Information Engineering, Henan University of Science and Technology, Luoyang 471023, China
Uav Path Planning Based On Improved Deep Deterministic Policy Gradients, Sen Zhang, Qiangqiang Dai
Journal of System Simulation
Abstract: Aiming at the problems of poor convergence and invalid exploration when UAVs perform path planning in complex environments, an improved deep deterministic policy gradient(DDPG) algorithm is proposed. Using a dual experience pooling mechanism to store success and failure experiences separately, the algorithm is able to use the success experience to strengthen the strategy optimization and learn from the failure experience to avoid the wrong path; an APF method is introduced to add a bootstrap term to the planning, which is combined with the exploration of noisy actions in a randomized sampling process to dynamically integrate the selected actions; multi-objective …
Optimization Of Vehicle Routing For Cross-Infection Risk In The Epidemic,
2025
School of E-commerce and Logistics Management, Henan University of Economics and Law, Zhengzhou 450046, China
Optimization Of Vehicle Routing For Cross-Infection Risk In The Epidemic, Xiaodong Shi, Yongcheng Guo, Mingqi Ma, Jiarui Pan
Journal of System Simulation
Abstract: In view of the safety risks associated with logistics distribution route optimization during public health emergencies, this paper investigates the vehicle routing problem by incorporating the risk of cross-infection, integrates the cross-infection risk caused by logistics activities in the epidemic area into the logistics distribution model, and establishes a logistics vehicle distribution model with the goal of cross-infection risk and cost. An improved genetic algorithm is designed for model optimization and solution. Based on the integration of chaos initialization population and adaptive crossover and mutation operations, a neighbor exclusion operator is further proposed to enhance the global search ability …
Remote Sensing Small Object Detection Based On Cross-Stage Two-Branch Feature Aggregation,
2025
College of Information Science and Technology, Qingdao University of Science and Technology, Qingdao 266061, China
Remote Sensing Small Object Detection Based On Cross-Stage Two-Branch Feature Aggregation, Jie Li, Yang Liu, Liang Li, Bengan Su, Jialong Wei, Guangda Zhou, Yanmin Shi, Zhen Zhao
Journal of System Simulation
Abstract: Aiming at YOLOv8's leakage and false detection problems caused by target scale difference and complex background in remote sensing small target detection, this paper proposes a remote sensing image small target detection method based on cross-stage two-branch feature aggregation. The global shared weights in the convolution operator and the context-aware weights of specific tokens in the attention are fused to obtain high-frequency local information and low-frequency global information; the global remote dependencies are captured using a lightweight MLP, and the parallel cross-stage learnable vision center mechanism is designed to capture the information of the local corner regions of the …
A Radar Countermeasure Modeling Method Incorporating Cognitive Bias,
2025
Rocket Force University of Engineering, Xi'an 710025, China
A Radar Countermeasure Modeling Method Incorporating Cognitive Bias, Rui Wang, Xiangyang Li, Dong Wang, Hongguang Ma, Zhili Zhang
Journal of System Simulation
Abstract: Cognitive bias, stemming from electronic measurement error and variability in human perception, exists in cognitive electronic warfare and affects the outcomes of conflicts. In this paper, the dynamic game approach is employed to develop a model for cognitive bias induced by incomplete information and measurement errors in cognitive radar countermeasures. The payoffs for both parties are calculated using the radar's anti-jamming strategy matrix A and the jammer's jamming strategy matrix B. With perfect Bayesian equilibrium, a dynamic radar countermeasure model is established, and the impact of cognitive bias is analyzed. Drawing inspiration from the cognitive bias analysis method used …
Digital Twin Framework For The Generation And Optimization Of Security Policies For Tsn Industrial Control Systems,
2025
School of Artificial Intelligence and Automation, Huazhong University of Science and Technology, Wuhan 430074, China
Digital Twin Framework For The Generation And Optimization Of Security Policies For Tsn Industrial Control Systems, Huimai Zhang, Xiaoya Hu, Chunjie Zhou
Journal of System Simulation
Abstract: The characteristic of multi-service flow integration in TSN industrial control systems makes it very difficult to establish an accurate mathematical model. In order to ensure the coordination between the security policy and the real-time operation of the system, a four-layer double-closed-loop digital twin framework of "physical layer-data layer-twin layer-service layer" serving the generation and optimization of security policies is proposed. The optimal security policy generation is achieved through the internal closed loop composed of iterative optimization between the initial security policy generation at the service layer and the deployment verification at the twin layer. The deterministic communication process between …
A Transfer Learning-Based Hybrid Model For Pm2.5 Concentration Prediction,
2025
School of Artificial Intelligence and Automation, Hohai University, Nanjing 211100, China
A Transfer Learning-Based Hybrid Model For Pm2.5 Concentration Prediction, Xinbiao Lu, Chunlin Ye, Yisen Chen, Wen Wu, Yudan Chen
Journal of System Simulation
Abstract: In order to solve the problems of increased computational cost due to irrelevant features and decreased prediction accuracy due to the difference in probability distribution caused by the change of data distribution over time in PM2.5 concentration prediction, this paper constructs a hybrid deep learning model TraTCN-LSTM-BiGRU based on migration learning. The meteorological factors related to PM2.5 concentration are selected as the model input using the mean-value heat map algorithm features; the source domain data and target domain data are divided by KL scatter and an adaptive layer is introduced into the model to achieve inter-domain distribution adaptation; the …
Trajectory Planning Of Quadruped Robot Over Obstacle With Single Leg Based On Deep Reinforcement Learning,
2025
College of Mechanical Engineering, Southwest Jiaotong University, Chengdu 610031, China
Trajectory Planning Of Quadruped Robot Over Obstacle With Single Leg Based On Deep Reinforcement Learning, Min Li, Sen Zhang, Xiangguang Zeng, Gang Wang, Tongwei Zhang, Dijie Xie, Wenzhe Ren, Tao Zhang
Journal of System Simulation
Abstract: Aiming at the problems of joint vibration and high energy consumption of quadruped robot in the process of walking over obstacles, a foot trajectory planning method of quadruped robot based on deep reinforcement learning SAC algorithm is proposed. Based on robot kinematics and Monte Carlo method, the motion space of the single-legged foot of quadruped robot is analyzed. A compound seventhdegree polynomial trajectory of the quadruped robot is planned. The SAC algorithm is used to train and obtain the low energy consumption obstacle crossing strategy of four-legged robot under different obstacle environment. The simulation results show that the compound …
