Projects

It has been critical and important for us to explore newer and innovative methods to interact, coordinate and control as modern living has been more complex and insecure due to the hostile environments. Even though the technological advances in the modern age has become considerably significant, major threats and issues are being dealt by us from non-traditional sources. The design and development of intelligent agent controllers that are adaptive and dynamic based on a novel concept of intelligent supervisory loop is the objective of this research. Research Objective #1: Architectural design and analysis of supervisory loop based multi-agent system for coordinated control, dynamics and planning. Research Objective #2: Develop an expert domain for knowledge generation and interaction in a multi-agent system. Research Objective #3: Develop an embedded systems and sensory networks using communication and control for the intelligent coordination and control of multiple agents.

This project focuses on designing and developing scalable intelligent adaptive control algorithms for renewable energy and distributed power system control. The main objective is to evaluate intelligent controllers designed based on a unique concept of supervisory loop principle for the control of fuel cell and micro-turbine distributed generation. A grant proposal to this effect has been submitted to NSF.

Currently in the process of designing an implicit model based intelligent controller for the control of functionally unknown, parametrically uncertain and/or multimodal systems, incorporating intelligent supervisory loop which can track changes in the system model. A grant proposal has been awarded to this effect as a part of summer research grant. External funding opportunities are being explored.

Currently working on designing, modeling and developing intelligent controllers to meet the following objectives: 1) Control complex systems based on a novel supervisory loop approach, 2 a) Design a global action evaluation framework in order to establish a modular control environment, b) Formulate and develop autonomous agents for modular interaction, 3) Develop an embedded system technology using network and microcontrollers and test these design formulations on specific practical applications. A grant proposal has been awarded to this effect as a part of summer research grant. External funding opportunities are being explored.

This research develops an approach for Voltage Stability Assessment and Improvement using MLP based ANN. The ANN with the capability of learning of the Voltage stability Index L is designed based on specific predefined input vector. The trained network is then tested for IEEE 30 bus system and a practical 260 bus system showing the signifance of the proposed technique.

A realtime DC Motor speed control using C505C microcontroller is developed in this project. The development environment is by using DAVE and Phytec Micro Vision packages and the module is tested using a DC Motor model.

This project was to develop a heuristic based game playing environment for human to human and human to machine Baroque Chess Game using LISP. The algorithm is used and troubleshooted for the various movements and for heuristic capability.

In this research I have developed a Self Tuning Regulator (STR) based parallel adaptive controller for the position tracking of permanent magnetic stepper motor (PMSM). The developed control algorithm is tested on a nonlinear PMSM model.

This research focus on developing a Fuzzy Logic based Model Reference Adaptive Controller suitable for control of dynamic Multi Modal System. The ability of Fuzzy systems to capture a nonlinear mapping between the input and output is explored inorder to change the desired dynamic reference model of a dynamic system looking at the auxiliary measurements. An adaptive controller is further used to control the plant output.

This research project focus on developing an Online dynamic neural network for the nonlinear functional approximation. A Radial Basis Function Neural Network (RBFNN) algorithm is developed for this purpose. The features of the algorithm is its dynamic capability and the ability to deal with "Dimensionality Problem". The developed algorithm is used learn nonlinear system and applied as an Intelligent Loop along with an Adaptive Controller for functional uncertainity.

One of the important aspect being addressed in this research project is to develop a neuro-fuzzy domain such that the functional and parametric uncertain system can be controlled effectively with this Intelligent Adaptive Control. A neural network algorithm working in parallel with the adaptive controller is the back bone of the system.

A new class of Intelligent Adaptive Control Designs for complex and multimodal dynamic systems is introduced. Those systems under consideration include dynamic systems that show multi-modality, scheduled and unscheduled ‘Jumps’, as well as unmodeled dynamics, and are often under the challenge of unforeseen changes due to plant dynamics and/or external influences. The focus of the dissertation is to develop Intelligent Supervisory Loops augmenting a stable Direct Model Reference Adaptive Controller in order to control such systems. In view of the above, four design formulations are developed which evolved from different methods of conceiving Intelligent Supervisory Loops. Subsequently, those formulations are structured into intelligent control algorithms and investigated with comprehensive simulation models of a Single Link Flexible Robotic Manipulator as well as a 6-DOF F16 Fighter Aircraft. In order to facilitate these dynamic systems variations, the control algorithms are applied on the developed models that are activated under artificially created structural fluctuations.