Computer Sciences Commons^™

Study On Task Allocation Of Uav Swarm Based On Cognitive Control, Ruixuan Wei, Zichen Wu

Journal of System Simulation

Abstract: A centralized allocation method for Unmanned Aerial Vehicle (UAV) swarm real-time task allocation is proposed. Based on the real-time battlefield situation, this method relies on the coordination scheduling layer to make allocation decisions, generates the strike order according to the strike efficiency ratio, and makes rolling optimization for real-time battlefield situation, so that the allocation results can always maintain the balance between optimization and efficiency within the current cognitive range. Particle swarm optimization (PSO) is used to solve the task assignment problem. In particular, a 2-D particle 0-1 coding and correction method for nonstandard particles is designed. The simulation …

Simulation And Optimization Of Supersonic Drag Characteristics Of Blunt Cone Ascender, Xiazhen Liu, Yuan Wu, Li Qi, Zhao Rui, Zhang Jian, Zhonghua Lu

Journal of System Simulation

Abstract: Reducing shock drag is an important objective of the aerodynamic optimization of Mars ascenders. The CFD(Computational Fluid Dynamics) method is used to study the supersonic drag characteristics of a typical blunt cone ascender. Under the design constraints of slenderness ratio of 0.42, 13 groups busbar shape set of zero-lift drag are compared, and the longitudinal static and dynamic stability are discussed. The research results show that, under the design constraints, the blunt shape with large bulbous and a gently sloped conical surface has low drag characteristics. The reason is that, after passing through the bulbous, the supersonic airflow …

Uniform Experimental Design Of Constrained Region Based On Evolutionary Algorithm, Jianing Wei, Hao Hao, Qutong Chang, Lin Tao, Zhang Hu

Journal of System Simulation

Abstract: The parameters of a simulation system are usually generated by the experimental design. Aiming at high design difficulty and computational cost of the uniform experimental design, of the constraint region, a two-stage differential evolutionary algorithm is further improved. The design is modeled as a constrained optimization problem. A strategy combining distribution estimation algorithm (EDA) and differential evolution (DE) is adopted. A point-deletion method is proposed to reduce the time complexity of optimizing the population uniformity. To demonstrate the advantages, the test instances and engineering applications are used in experimental analysis. The experimental results show that the performance, stability, and …

Study On Interior Space Pedestrian Evacuation Model Elite Chaos Search Strategy, Wei Juan, Zhongyu Li, You Lei, Yangyong Guo, Zhihai Tang, Zhouyi Hu

Journal of System Simulation

Abstract: In order to improve traditional field model of easily falling into the queue problem during simulating the evacuation of dense crowds, an improved pedestrian evacuation model in interior space is proposed based on the elite chaos search strategy. The pedestrian mobile income at each moment is calculated in combination with the field value, capacity and average speed, and a field model for pedestrian evacuation is presented. On this basis, the objective optimization function of the minimum evacuation time and minimum queue length is given, and the elite chaos search strategy is used to achieve the above objective function solution, …

Fast Alignment Of Bim Products Based On Structure Matching, Xiaojun Liu, Changyan He, Liu Chang, Jinyuan Jia

Journal of System Simulation

Abstract: The fast alignment of BIM products based on structure matching is realized. An Item-based Structure Graph (ISG) is proposed as the structure-aware shape descriptor for BIM products. An improved graph matching algorithm upon Factorized Graph Matching (FGM) is implemented in ISG matching, in which the mismatch caused by symmetry switching is avoided. Further extension is made to support the alignment under partial matching for BIM products from the same family. The experiment result shows that ISG can meet the characteristics of BIM products, improve the efficiency and accuracy, and can be applied in dealing with products alignment, BIM data …

Position And Attitude Estimation Based On Combination Matching In Calibration Area, Cai Peng, Chaoping Shen, Hongyan Li

Journal of System Simulation

Abstract: Vision navigation technology of scene matching needs hardware to measure the distance and attitude of the camera. A method of position and attitude estimation based on the combination of feature points in the calibration area is proposed. The software can get the camera position and attitude of the real-time image by selecting the best set of Scale-Invariant Feature Transform(SIFT) feature matching points of the reference image calibration area of the real-time image, and calculating the local coordinates of the ground of SIFT feature matching points based on the triangle internal linear interpolation method, using space resection, avoids the defect …

Numerical Simulation Of Aerodynamic Characteristics For Double-All-Wing Unmanned Aerial Vehicle Based On Computational Fluid Dynamics Theory, Kuizhi Yue, Zhang Yuan, Liangliang Cheng, Dazhao Yu

Journal of System Simulation

Abstract: Follow the rapid progress of the nation's blue ocean strategic projection capability of sea power, an Unmanned Aerial Vehicle (UAV) with double-all-wing configuration is studied to match the trend. The UAV is conceptually designed by CATIA software, the aerodynamic characteristic of the model is numerically simulated by computational fluid dynamics (Computational Fluid Dynamics, CFD) theory, and the performance of the UAV to take off from the aircraft carrier is also analyzed. The lift coefficient, drag coefficient and lift-drag ratio of the double-all-wing UAV model with a angle of attack of 0 can reach 0.596 2, 0.037 6 and 15.824 …

High Adaptability Vehicle Network Communication Protocol Based On Aodv, Wang Tong, Gao Shan, Tianshu Dai

Journal of System Simulation

Abstract: Routing protocol is an essential part of vehicular ad-hoc network (VANET). According to the characteristics of frequent changes of vehicle node location in urban environment, an AODV routing protocol based on Network Topology Changes (NTCAODV) is proposed. According to the speed and relative position information, by calculating the stability coefficient of nodes, the link connection is established on the basis of principle of "optimal stability", and the idea of extended loop is improved according to recorded the link establishment time. Simulation results show that NTCAODV routing protocol has good performance in various urban environments, improves …

Evaluation Method On Multi-Stage Bayes Information Fusion For Scatters Hwil Simulation, Yecheng Wang, Ma Jing, Congjun Jin, Jie Li, Bin Shi

Journal of System Simulation

Abstract: In order to simulate the complex 3D target in the RF HWIL (hardware in the loop simulation) system, the equivalence framework of multiple scattering centers of complex 3D target for three stages of outfield test data, mathematics simulation data and HWIL data are introduced. Using the pre-test information as each stage for transmission, evaluation method is designed by multi-stage Bayes information fusion for HWIL simulation. Based on prior information of outfield test data and mathematics simulation data, by using Bayes information fusion theory, the average data reconstruction of each stage is obtained and compared and high reliability …

Collaborative Optimization Of Ship Stowage Plan And Yard Container Unloading For Container Terminal, Zhixiong Liu, Hanwen Qian, Jialan Yan

Journal of System Simulation

Abstract: Stowage plan is an important component of the operation production for the container terminal, which directly affects the handling efficiency and shipping safety. The aim is minimizing the total reloading numbers of both the yard and the ship bay, in view of the constraint of ship heeling moment and loading for light upon heavy, the model is employed for collaborative optimization of stowage plan and yard unloading for multiple destination ports and multiple bays. The hybrid evolutionary strategy algorithm is adopted with the local search algorithm. As to ship stowage play for multiple bays, a heuristic rule for the …

Simulation Evaluation For Service Facilities Distribution And Flow Lines Optimization On Urban Rail Transfer Station, Liu Feng, Chunfu Shao, Hongfei Jia, Chunjiao Dong

Journal of System Simulation

Abstract: The rationality of passenger flow lines design and service facilities distribution inside urban rail transfer station directly affects the passenger's transfer efficiency. The Simulation technology are used to evaluate the station's transfer efficiency, and the optimal plan can be proposed, which can not only avoid the crowded passenger flow, but also improve the efficiency. The features of peak passenger flow are considered, the service level of the passenger's stay time is given, the change of congested area's passenger flow density is expressed in the form of time-varying chart. Combined with the overall chart of the average passenger flow density, …

Transmission Process Prediction Of Novel Coronavirus Based On System Dynamics, Xuepeng Lu, Shang Jiao, Junhui Zhao, Lulu Lü, Zhou Li

Journal of System Simulation

Abstract: The transmission characteristics of novel coronavirus is considered and a new SE4IR2 model based on the principle of system dynamics is proposed. The US epidemic data from June to November is used to set the isolation rate and other parameters, and the SE4IR2 model is used to fit, analyze and predict the development of the epidemic trend in the next stage. The empirical part uses the data from June to November in the United States to achieve the parameters of the SE4IR2 model, obtains the parameter values in December and January through the time series prediction model, and compares …

Unsupervised Anomaly Instance Segmentation For Baggage Threat Recognition, Taimur Hassan, Samet Akçay, Mohammed Bennamoun, Salman Khan, Naoufel Werghi

Computer Vision Faculty Publications

Identifying potential threats concealed within the baggage is of prime concern for the security staff. Many researchers have developed frameworks that can automatically detect baggage threats from security X-ray scans. However, to the best of our knowledge, all of these frameworks require extensive training efforts on large-scale and well-annotated datasets, which are hard to procure in the real world, especially for the rarely seen contraband items. This paper presents a novel unsupervised anomaly instance segmentation framework that recognizes baggage threats, in X-ray scans, as anomalies without requiring any ground truth labels. Furthermore, thanks to its stylization capacity, the framework is …

On Communication For Distributed Babai Point Computation, Maiara F. Bollauf, Vinay A. Vaishampayan, Sueli I.R. Costa

Publications and Research

We present a communication-efficient distributed protocol for computing the Babai point, an approximate nearest point for a random vector X∈Rn in a given lattice. We show that the protocol is optimal in the sense that it minimizes the sum rate when the components of X are mutually independent. We then investigate the error probability, i.e. the probability that the Babai point does not coincide with the nearest lattice point, motivated by the fact that for some cases, a distributed algorithm for finding the Babai point is sufficient for finding the nearest lattice point itself. Two different probability models for X …

Mathematical Optimization Algorithms For Model Compression And Adversarial Learning In Deep Neural Networks, Tianyun Zhang

Dissertations - ALL

Large-scale deep neural networks (DNNs) have made breakthroughs in a variety of tasks, such as image recognition, speech recognition and self-driving cars. However, their large model size and computational requirements add a significant burden to state-of-the-art computing systems. Weight pruning is an effective approach to reduce the model size and computational requirements of DNNs. However, prior works in this area are mainly heuristic methods. As a result, the performance of a DNN cannot maintain for a high weight pruning ratio. To mitigate this limitation, we propose a systematic weight pruning framework for DNNs based on mathematical optimization. We first formulate …

Treadmill Assisted Circumvention Of Wearable Sensors-Based Gait Authentication, Rajesh Kumar

Dissertations - ALL

Wearable sensor-based gait patterns are considered a promising means for future authentication systems. This dissertation examines whether circumvention of such systems can be accomplished by imitating sensor readings and providing external mechanical support. The specific machine that we used in the experiment was a digital treadmill, which provides a suitable platform for the human imitators to control, adjust and adapt several factors, such as speed, step-length, step-width, and thigh-lift that affect sensor readings. Moreover, it was easy for imitators to remember the gait factors' specific levels and repeat the learned pattern on-demand over a treadmill.

Two novel imitation-based attacks are …

Multi-Modal Self-Supervised Representation Learning For Earth Observation, Pallavi Jain, Bianca Schoen Phelan, Robert J. Ross

Conference papers

Self-Supervised learning (SSL) has reduced the performance gap between supervised and unsupervised learning, due to its ability to learn invariant representations. This is a boon to the domains like Earth Observation (EO), where labelled data availability is scarce but unlabelled data is freely available. While Transfer Learning from generic RGB pre-trained models is still common-place in EO, we argue that, it is essential to have good EO domain specific pre-trained model in order to use with downstream tasks with limited labelled data. Hence, we explored the applicability of SSL with multi-modal satellite imagery for downstream tasks. For this we utilised …

Real-Time Adaptive Sensor Attack Detection And Recovery In Autonomous Cyber-Physical Systems, Francis Akowuah

Dissertations - ALL

Cyber-Physical Systems (CPS) tightly couple information technology with physical processes, which rises new vulnerabilities such as physical attacks that are beyond conventional cyber attacks.Attackers may non-invasively compromise sensors and spoof the controller to perform unsafe actions. This issue is even emphasized with the increasing autonomy in CPS. While this fact has motivated many defense mechanisms against sensor attacks, a clear vision of the timing and usability (or the false alarm rate) of attack detection still remains elusive. Existing works tend to pursue an unachievable goal of minimizing the detection delay and false alarm rate at the same time, while there …

Augmented Human Machine Intelligence For Distributed Inference, Baocheng Geng

Dissertations - ALL

With the advent of the internet of things (IoT) era and the extensive deployment of smart devices and wireless sensor networks (WSNs), interactions of humans and machine data are everywhere. In numerous applications, humans are essential parts in the decision making process, where they may either serve as information sources or act as the final decision makers. For various tasks including detection and classification of targets, detection of outliers, generation of surveillance patterns and interactions between entities, seamless integration of the human and the machine expertise is required where they simultaneously work within the same modeling environment to understand and …

An Economical Method For Securely Disintegrating Solid-State Drives Using Blenders, Brandon J. Hopkins Phd, Kevin A. Riggle

Journal of Digital Forensics, Security and Law

Pulverizing solid-state drives (SSDs) down to particles no larger than 2 mm is required by the United States National Security Agency (NSA) to ensure the highest level of data security, but commercial disintegrators that achieve this standard are large, heavy, costly, and often difficult to access globally. Here, we present a portable, inexpensive, and accessible method of pulverizing SSDs using a household blender and other readily available materials. We verify this approach by pulverizing SSDs with a variety of household blenders for fixed periods of time and sieve the resulting powder to ensure appropriate particle size. Among the 6 household …

Learning To Learn Variational Semantic Memory, Xiantong Zhen, Yingjun Du, Huan Xiong, Cees G.M. Snoek, Ling Shao

Machine Learning Faculty Publications

In this paper, we introduce variational semantic memory into meta-learning to acquire long-term knowledge for few-shot learning. The variational semantic memory accrues and stores semantic information for the probabilistic inference of class prototypes in a hierarchical Bayesian framework. The semantic memory is grown from scratch and gradually consolidated by absorbing information from tasks it experiences. By doing so, it is able to accumulate long-term, general knowledge that enables it to learn new concepts of objects. We formulate memory recall as the variational inference of a latent memory variable from addressed contents, which offers a principled way to adapt the knowledge …

Representation Of Nonlinear Pseudo-Random Generators Using State-Space Equations, Raghad K. Salih

Emirates Journal for Engineering Research

The idea of research is a representation of the nonlinear pseudo-random generators using state-space equations that is not based on the usual description as shift register synthesis but in terms of matrices. Different types of nonlinear pseudo-random generators with their algorithms have been applied in order to investigate the output pseudo-random sequences. Moreover, two examples are given for conciliated the results of this representation.

Structured Latent Embeddings For Recognizing Unseen Classes In Unseen Domains, Shivam Chandhok, Sanath Narayan, Hisham Cholakkal, Rao Muhammad Anwer, Vineeth N. Balasubramanian, Fahad Shahbaz Khan, Ling Shao

Computer Vision Faculty Publications

The need to address the scarcity of task-specific annotated data has resulted in concerted efforts in recent years for specific settings such as zero-shot learning (ZSL) and domain generalization (DG), to separately address the issues of semantic shift and domain shift, respectively. However, real-world applications often do not have constrained settings and necessitate handling unseen classes in unseen domains – a setting called Zero-shot Domain Generalization, which presents the issues of domain and semantic shifts simultaneously. In this work, we propose a novel approach that learns domain-agnostic structured latent embeddings by projecting images from different domains as well as class-specific …

Ai Output: A Human Condition That Should Not Be Protected Now, Or Maybe Ever, Xiao Wang

Chicago-Kent Journal of Intellectual Property

AI is usually considered to be a form of automatic and autonomous work, but when applied to the creation of literary and artistic works, challenges arise in deciding whether the AI is the de facto author of its output and whether AI outputs or AI-generated products should be protected under the copyright system. This article argues that these outputs should be human creations because the working principles of AI determine that AI functions merely as a mathematical tool applied by humans to not only conceive of but also to execute the creation of AI outputs. The creativity reflected in these …

Graph-Theoretic Partitioning Of Rnas And Classification Of Pseudoknots-Ii, Louis Petingi

Publications and Research

Dual graphs have been applied to model RNA secondary structures with pseudoknots, or intertwined base pairs. In previous works, a linear-time algorithm was introduced to partition dual graphs into maximally connected components called blocks and determine whether each block contains a pseudoknot or not. As pseudoknots can not be contained into two different blocks, this characterization allow us to efficiently isolate smaller RNA fragments and classify them as pseudoknotted or pseudoknot-free regions, while keeping these sub-structures intact. Moreover we have extended the partitioning algorithm by classifying a pseudoknot as either recursive or non-recursive in order to continue with our research …

A Deep Learning Approach For Forecasting Global Commodities Prices, Ahmed Saied Elberawi, Mohamed Belal Prof.

Future Computing and Informatics Journal

Forecasting future values of time-series data is a critical task in many disciplines including financial planning and decision-making. Researchers and practitioners in statistics apply traditional statistical methods (such as ARMA, ARIMA, ES, and GARCH) for a long time with varying accuracies. Deep learning provides more sophisticated and non-linear approximation that supersede traditional statistical methods in most cases. Deep learning methods require minimal features engineering compared to other methods; it adopts an end-to-end learning methodology. In addition, it can handle a huge amount of data and variables. Financial time series forecasting poses a challenge due to its high volatility and non-stationarity …

Entity Retrieval Using Fine-Grained Entity Aspects, Shubham Chatterjee, Laura Dietz

Computer Science Faculty Research & Creative Works

Using entity aspect links, we improve upon the current state-of-the-art in entity retrieval. Entity retrieval is the task of retrieving relevant entities for search queries, such as "Antibiotic Use in Livestock". Entity aspect linking is a new technique to refine the semantic information of entity links. For example, while passages relevant to the query above may mention the entity "USA", there are many aspects of the USA of which only few, such as "USA/Agriculture", are relevant for this query. By using entity aspect links that indicate which aspect of an entity is being referred to in the context of the …

A Health Elearning Ontology And Procedural Reasoning Approach For Developing Personalized Courses To Teach Patients About Their Medical Condition And Treatment, Martin Michalowski, Szymon Wilk, Wojtek Michalowski, Dympna O'Sullivan, Silvia Bonaccio, Enea Parimbelli, Marc Carrier, Grégoire Le Gal, Stephen Kingwell, Mor Peleg

Articles

We propose a methodological framework to support the development of personalized courses that improve patients’ understanding of their condition and prescribed treatment. Inspired by Intelligent Tutoring Systems (ITSs), the framework uses an eLearning ontology to express domain and learner models and to create a course. We combine the ontology with a procedural reasoning approach and precompiled plans to operationalize a design across disease conditions. The resulting courses generated by the framework are personalized across four patient axes—condition and treatment, comprehension level, learning style based on the VARK (Visual, Aural, Read/write, Kinesthetic) presentation model, and the level of understanding of specific …

Decoding Clinical Biomarker Space Of Covid-19: Exploring Matrix Factorization-Based Feature Selection Methods, Farshad Saberi-Movahed, Mahyar Mohammadifard, Adel Mehrpooya, Mohammad Rezaei-Ravari, Kamal Berahmand, Mehrdad Rostami, Saeed Karami, Mohammad Najafzadeh, Davood Hajinezhad, Mina Jamshidi, Farshid Abedi, Mahtab Mohammadifard, Elnaz Farbod, Farinaz Safavi, Mohammadreza Dorvash, Shahrzad Vahedi, Mahdi Eftekhari, Farid Saberi-Movahed, Iman Tavassoly

Publications and Research

One of the most critical challenges in managing complex diseases like COVID-19 is to establish an intelligent triage system that can optimize the clinical decision-making at the time of a global pandemic. The clinical presentation and patients’ characteristics are usually utilized to identify those patients who need more critical care. However, the clinical evidence shows an unmet need to determine more accurate and optimal clinical biomarkers to triage patients under a condition like the COVID-19 crisis. Here we have presented a machine learning approach to find a group of clinical indicators from the blood tests of a set of COVID-19 …

Efficient Metadata Lookup In Inline Deduplication Systems Leveraging Block Similarity, Rakesh Gururaj

Master's Projects

Data deduplication is a concept of physically storing a single instance of data by eliminating redundant copies to save the storage space. The adoption of deduplication is minimal in actively accessed primary storage because of its complexities, such as random access patterns to data and the need for quicker request response time. Most of the solutions designed for primary storage are offline and dependent on the concept of locality. This paper proposes an inline deduplication system with a Machine Learning based cache eviction policy to reduce the metadata overhead in the deduplication process, eliminate the redundant writes and improve the …