Towards Scientific and Technological Advancements inmodern human history, Artificial Intelligence (AI) has achieved a profound position in the world of Arts, Science, and Technology. Rexsy has been involvedin AI Systems Design and Development since the 80's, thus,various AI discussions are presentedhere for understandings and knowledge acquisitions.
<> REXSY UNIVERSAL COSMIC MUSIC <>
Einstein, Zadeh, Newton, Neils Bohr,
Darwin, Da Vinci, Simon Haykin,
Galileo, Planck, Kurzweil, Kagen Atkinson,
Carl Sagan, Lennart Ljung, Tesla, Maxwell, Faraday,
Benjamin Franklin, Stephen Hawking, Edison, Descartes!
Today, AI can be found in the world of Scientific ResearchandEngineering in Advanced Stochastic Adaptive Signal/Image Processing,NeuralFuzzy Adaptive Systems, Ontology & Expert Systems, Pattern Recognition, Robotic/Computer Vision, Data Mining, Knowledge Discovery, Algorithms Development, Financial Analytics Engineering, Biometrics Security System Design, Adaptive IntelligentControls, RF Communications, Radars, Avionics, Multi-Agent Complex Adaptive Systems, Evolutionary Systems, and Defense Systems Engineering.
On World Applications, Rexsy has beeninvolved withthe following technologies:
As a Scientist & Engineering Consultant Expert, Rexsy has been involoved in the following AI projects:
Design and Development of Advanced Evolutionary Multi-Agent Expert Systems (AEXSYS) based on Adaptive Recurrent Fuzzy Neural Networks, State Space Innovations Models, Multi-variable Nonlinear Autoregressive Moving Average (ARMAX) Systems, Advanced Pattern Recognition Theory, Ontological Expert Systems, and Generalized Hybrid AI System Architectures (Series/Series, Series/Parallel, etc.). These AEXSYS Systems possess both Fuzzy Knowledge-Based and Computational Intelligence evolving through time towards optimal endeavors.
Latest Project involved Design & Development of REXSY Universal System (REXSYS), which is the Fusion and Integration of Advanced Complex Adaptive MultiAgent Ontological Expert Systems ACXSYS, ACXSYS.mix, ACXSYS.vsm, ACXSYS.nlp, ACXSYS.quantum, and ACXSYS.hybrid, in both Classical, Quantum, and hybrid Classical/Quantum computing environments, for both centralized Web-based and decentralized Blockchain applications. This is the universal Rexsys Quantum AI Systems (REXSYS).
Recent projects included design and development of Web Data Detection, Extraction, and Mining for Business application, based on Artificial Intelligence (AI) technologies. The novel approach used Vision-based AI techniques for detecting, extracting, and processing Data of interest in noisy stochastic environment. Also, Designed and Developed Advanced Relevant Data/Documents Search Systems based on the Evolutionary Vector Space Model (EVSM), Genetic Latent Semantic Analysis (GLSA), Neural Genetic Boolean Search (NGBS), K-means & Fuzzy C-means Clustering, and Neural Pattern Recognition/Search Engine Optimization (SEO) Algorithms. These systems showed great optimal performances, from integration of both Numerical/Digital and Linguistic Information Dynamics, and unification of both Computational and Knowledge-based intelligence. Ultimately, integration of the EVSM, GLSA, and NGBS constituted an extremely powerful Search System on the Internet World of Information.
For applications on Internet high-speed Data Transmission, participated in the Design and Development of Optimal Reference Web-Page/Document Compression based on the Vector Space Model (VSM) and Latent Semantic Analysis (LSA). Furthermore, Advanced versions of the above Systems were designed based on Users Predictive Patterns, which generated Optimal Reference Pages ahead of time for highly efficient Data Compression.
With respect to University Enrollment Management and Controls, Developed and Designed the Predictive Analytics Expert Systems, which utilized Neural Fuzzy Architecture for Students Predictve Enrollment Patterns, and with such Prediction Data available, the University Management would be able to effectively design Controls Strategies for Optimal Enrollments.
Designand Development ofthe Advanced Complex Adaptive MultiAgent Ontological Expert System (ACXSYS), which consisted of various groups of Intelligent Agents working and performing complex tasking/functions in parallel to achieve the final goals/endeavors. Responsibilities of the Agents included Massive Data Filtering/Mining, Clustering/Segmentation, Spectral Estimation, Blind Signals Separation/Deconvolution, Pattern Detection/Recognition, Future Events Prediction, Information Tracking, Expert Knowledge Mining/Discovery, Intelligent Decision Support, and Complex Adaptive Transactions. The technologies and systems architectures embedded in these Agents included the Evolutionary Fuzzy Neural Expert Systems, Genetic Neural Pattern Recognition, Fuzzy Learning Vector Quantization & C-means Clustering, Stochastic State Space Predictions, and Complex Adaptive Systems Theory. Also, Advanced Biometrics Pattern Segmentation, Detection, Tracking, and Recognition techniques were applied to the Intelligent Multi-Dimensional Adaptive Mining/Discovery process on Time-Series/Image data, as well as Stochastic Evolutionary Complex Adaptive Controls Expert Systems on Multi-Agents Optimizations and Controls.
Collaborated with high-tech companies on Research, Design, and Development of cutting-edge Technologies, and Solutions/Systems, and wrote world-class STTR/SBIR proposals for winning innovation research projects. Some of these projects included the architectural design of Multi-Agent Expert System for optimal performance of advanced Telerobotics Surgery with Multiseneory/Multimodal Interface. Another project involved Design and Development of the Multi-Agent Multi-Target Tracking System for immersive Virtual Reality (VR) environment for infantry battle training. With respect to Adaptive Flight Controls, Designed Intelligent Multi-Agent Flight Control Systems for effectively and robustly autonomous controls of UAVs with missions in challenging environmental /weather conditions.
Design ofthe OBX, which involveed Integrated fusion of Ontology and Advanced Data Mining System, with adaptive feedback capabilities for real-time Learning, Communications, and Controls. For the OBX, developed the Multi-Agent Complex Adaptive System Architecture, where the OBX was embedded in the agents intelligence and knowledge structure. In general, the key advantages of the OBX were the fusion power of both Computational Intelligence and Semantic knowledge-based Ontological Intelligence, on-line Self-Learning capabilities based on the closed-loop feedback structure, and the powerful Complex Adaptive System Architecture. The OBX, based onWordly Applications, successfully demonstrated its powerful capabilities.
Building an Ontology-based Federated Dynamic Data Distribution System. For the project, developed an overall system architecture based on the Multi-Agent System (MAS) structure, where the Agents consisted of an Ontology (O), Query Federator QF), Data Mining (DM), and Hierarchical Storage Management (HSM) subsystems. These Agents worked in concert under the management of an Intelligent Ontology Expert System for optimal dynamic data distribution. The Data Mining utilized technologies including Genetic Fuzzy Clustering, Neural Pattern Recognition, System Identification & Modeling, Hidden Markov Models, Support Vector Machines, Kernel Systems, and Evolutionary Systems theory. The Ontology integrated both Knowledge (Rules) and Computational Intelligence, and adaptively updated its structure online under the changing environmental dynamics. Also worked on the conceptualization of adaptive intelligent Ontology architectures, including the Evolutionary Neural Fuzzy Ontology System associated with the Multi-Agent Systems (MAS) theory.
Anotherproject included application of Multi-Agent Systems (MAS) in missile defense environment, where the agents (Trackers, Predictors, Discriminators, Interceptors, Controllers, Pattern Recognizers, Data Miners, etc.) interacted cooperatively to achieve overall system goals, through a communications/controls network mastered by an Expert System, where Evolutionary Algorithm architecturally took part of the overall controls, coordination, negotiation, and optimization. The co-evolution, fusion, and cooperation of the MAS and evolutionary Expert System lead to the development of the Multi-Agent Expert System (MAXSYS). The MAXSYS was applied to Radar Target Tracking, Prediction, Navigation, Guidance, Controls, and Interception, where the Interacting Multiple Model (IMM)/Extended Kalman Filter (EKF) agents were employed, and the simulation was developed for defense environments. In parallel with the Evolutionary Algorithm/System mentioned above, the Adaptive Evolutionary Annealing Simplex System was under consideration for incorporation into the MAXSYS.
Recently involved in the Research, Design and Development of the most advanced Radar Systems using Phased Array Antennas, which required elengantly sophisticated Adaptive Processing Algorithms for Objects location detection and tracking, and based on Advanced AI theories for the design of the Stochastic Neural Fuzzy System. The system, with parallel architecture, intelligently and adaptively processed Radar data in real time including fast training for updating its knowledge while optimizing the multi-objective fitness function for unknown multi-variables estimation, under extremely noisy environment. Also, a Stochastic Genetic Kalman System was designed and developed to support and enhance the Radar data processing capabilities upon achieving optimal performance.
For a defense project,performed research, development and design of advanced detection, tracking, and identification of objects using Cellular Neural Networks (CNN) as the front-end processor, which was interfaced with other Digital Signal Processing (DSP) and Artificial Intelligence (AI) modules including the Extended Kalman Tracking, Neural Fuzzy pattern Recognition, Adaptive Multi-Spectral Estimation, Signal/Noise Cancellation, Evolutionary Genetic Optimization, and Object Detection Systems. Also, designed algorithms directly with the CNN as a main central processor for various applications in Adaptive Equalization, System Identification, Noise Modeling, Interference Cancellation, Adaptive Filtering, and Pattern Recognition.
Performed studies and analysis on Radar Tracking Systems. Designed the EKF-based (Extended Kalman Filter) Interacting Multiple Model (IMM) tracking filter, which involved investigation of the state space dynamical models, formulation of models transformation and mixing, implementation of the IMM algorithm, and design of the Matlab simulation system. Also conducted research and development of interacting multiple model (IMM) Radar Tracking Systems based on Neural Extended Kalman Filtering (NEKF) architecture. Various NEKFs were explored to adaptively model system nonlinear time-varying characteristics, which included the combined and coupled state/weights system identification, dual EKF optimization algorithm for separated states and weights estimation, NEKF based on state space innovations models, and advanced neural optimization of the state space target tracking models.
For a SBIR project, designed an Advanced Evolutionary Algorithm for radar/sensor resources scheduling and management system for applications in Missile Defense. The development process included designing the encoding system based on symbolic mathematical expressions with a multi-objective fitness function. This system showed optimal performance in achieving the design objectives with fast convergence speed, and applicable to real-time implementation.
Researched, Developed, and Designed Advanced Artificial Intelligence (AI) Systems for applications in Security, Surveillance, Intelligence, Human Detection & Identification, Biometric Systems, Interactive Multimedia Data Processing, Automated Video Indexing, AI Video Compression, Robotic Vision, Knowledge Discovery, Data Mining, Communications, and Controls. Responsibilities included designing Smart Machine Vision, Audio-Visual Classification, Object Segmentation, Detection, Identification, Tracking, Recognition, and Expert Control Systems, based on Neural Networks Theory, Adaptive Fuzzy Clustering, Genetic Algorithmic Optimization, Principal and Independent Component Analysis (PCA & ICA), Multivariable Nonlinear Adaptive Signal Processing, Genetic Fuzzy Neural System Architecture, Time & Space Multi-dimensional Signal Processing, Stochastic Adaptive Evolutionary Systems, and AI Expert Systems Theory.
Developed A Generalized Stochastic Adaptive Pattern Recognition Expert System, which was based on Adaptive Signal Processing, Parametric Estimation, Data Prediction, Kalman Filtering, System Identification Theory, Object Detection & Tracking, Nonlinear Adaptive PCA & ICA, Fuzzy C-Means & Learning Vector Quantization, Feature Extraction Systems, Adaptive Nonlinear Evolutionary Feedback Controls, Recurrent Neural Fuzzy Systems, Probabilistic & Generalized Regression Radial Basis Networks, Statistical Data Processing, and Adaptive Expert Systems Theory. In addition, the system employed robust adaptive techniques such as the Stochastic Gradient, Newton, Minimum Variance, and Evolutionary Algorithms for System Optimization and Nonlinear Feedback Controls.
Researched, designed, and developed Telecommunication Systems including Modems V.17, V.22, V.27, V.29, V.32, V.34, and V.90. Responsibilities included advanced modeling of the above Communication Systems based on Matlab/Simulink as reference against the assembly code in the design and development process. In parallel with systems modeling, performed advanced algorithms analysis and optimization, modified DSP software, and organized existing software into the General Unified System Architecture. With respect to algorithms engineering, conducted research and analysis on Communication System Modulation, Demodulation, Convolutional Coding, Optimal Viterbi Decoding, Adaptive Equalization, Jitter Tracking and Carrier Phase Recovery Algorithm, Adaptive Timing Recovery System, and Echo Cancellation. Additionally, performed Stochastic Signal Processing, Advanced Detection & Estimation, Kalman Filtering, and Artificial Intelligence Designs in Communication Systems. Also, developed Matlab/Simulink library of systems models for the V.xx modems (above) and other signal processing modules.
Worked onadvanced Geophysical data processing projects in both R&D and production environments. Projects under management included the Electromagnetic, and Gravity Airborne Systems modeling, simulation, design, and development. As part of responsibilities, directed research on the Genetic Neural Fuzzy Expert System for multi-dimensional pattern recognition which involved symbolic/subsymbolic data processing approaches. Additionally, some of the advanced technologies employed in various technical projects included the unsupervised blind deconvolution, separation, estimation, system identification, and pattern recognition. These technologies were based on the independent component analysis (ICA ), principal component analysis (PCA), and fuzzy neural adaptive systems theory.
Performed research, design, and development of digital signal processing systems for applications in Geophysics. Projects included developing advanced DSP algorithms for adaptive spectral estimation, signal modeling, adaptive noise cancellation, system identification, and adaptive filtering for processing exploration data. These advanced algorithms were based on the stochastic system identification and prediction theory, adaptive neural networks, and fuzzy inference systems. For these applications, signal and system model development included the multivariable MA, ARX, ARMAX, general linear, state space innovations, and neural fuzzy architectures. Also, for estimating and training these models, the minimum variance, stochastic gradient, Newton Optimization, and genetic algorithms were among some of the optimization methods used in the design process. The above tasks were performed interactively in both time and frequency domains.
Designed and developed advanced signal processing and control algorithms for World Financial Systems. Theories based on Expert Systems, Stochastic Processes, Fuzzy Controls, Neural Signal Processing, and Genetic Algorithms were employed in real-time processing of financial data upon optimizing investment strategies. Systems design included Adaptive Neural Fuzzy Signal Prediction, Stochastic Spectral Estimation, Time-Frequency Signal Analysis, Adaptive Blind Separation and Deconvolution, Cycles Analysis, Multivariable Filtering Systems, Pattern Recognition, and Stochastic Principal Component Analysis. The above System Design possessed a generalized architecture which could be utilized in various applications in Engineering, Geophysics, Defense, and Science. Multivariable inputs to the system included world market indexes, analyzed stock prices, bond yields, interest rates, commodity prices, and global intermarket data. System sampling rates may vary from minutes, hours, days, weeks, months, and quarters, to years of historical dynamics.
Designed and developed control strategies for water treatment systems. Projects included design of Expert Systems in parallel with Neural Networks and Fuzzy Logic for controlling nonlinear time varying processes; various control and signal processing modules were developed and linked to the Expert System including system identification, digital controls (PID, Logic, Adaptive, etc.), statistical signal processing, mathematical optimization, and failure detection algorithms. In the analysis and design process, responsibilities also required systems modeling, simulation, and optimization for achieving the best system design. Works were performed on the VAX/VMS (Supervisory Control using Plant Information (PI) system), ABB-MOD300 (Distributed Control System), and PCs (systems simulation and PLC programming). In parallel with controls engineering, also performed digital signal processing tasks including stochastic signal analysis, estimation, and prediction to complement control strategies. Also, organized and coordinated technical training courses and seminars for the Controls System Department at the plant.
Please contact Rexsy if you need Engineering Consulting on AI Systems Research, Design & Development.