Ashwin Sundarka

Topic: Nonlinear system Identification and Control of an SCR system Neural Networks a databased modelling approach is used for identification. Keeping the importance of data, Design of Experiment(DoE) is used to create the training data. One of the applications of the project is to use the model in the ECU, hence the needed experiments are conducted. For control, PI tuned with Evolutionary algorithm is used. Two different configurations of PI is taken over here, single PI & gain-scheduled PI.

Project here focuses on the development of different Collison avoidance algorithms and reference tracking controllers for an underwater vehicle. ‘BlueRov2 Heavy configuration' is used as plant model. Local collision avoidance algorithms, VFH+ and VFH Star are developed for faster real-time implementation. For reference tracking, two controller structures have taken here. One is to control velocity and the other one is for position of system. Both structures are performed with LQR and H∞ controller

Responsible for taking exercise classes of 'Control Systems Theory and Design', which includes both conceptual problems and programming in MATLAB. Also, have conducted a workshop on Mini Segway, where we controlled a single wheeled segway by designing LQR and LQG based controller.

Note: 1.6 Specialisation in Control Systems & Identification Major Subjects: Control System Theory and Design, Optimal and Robust Control, Linear and Nonlinear System Identification, Applied Humanoid Robotics, Mechatronics Systems, Nonlinear Dynamics, and Embedded Systems

Note: 1.2 Subjects covered in the area of Mechanics, Electronics, Electricals, Industrial Automation