Semi-Plenary Speakers

Semi-Plenary Speakers

The conference is happy to present four semi-plenary talks by eminent people from industry and academia on traditional and modern applications of control engineering.

  1. Prof. Maruthi R. Akella, E.P. Schoch Endowed Professor, University of Texas, Austin

    Reliable Autonomy for Aerospace Robotics: Perception, Learning and Trust
    Abstract: Many future aerospace and robotic missions are expected to be fielded across highly uncertain operating environments thereby limiting the possibility for online command and control from ground stations. The complex interplay between autonomy and onboard decision support systems introduce new vulnerabilities that are extremely hard to predict with most existing guidance and control tools. Perception, learning, and trust can be generally viewed to characterize autonomy as overarching system-level properties. In this lecture, we review some recent theoretical advances in learning-oriented and information-aware path-planning, covariance control, and non-myopic sensor scheduling approaches for autonomous systems. To rigorously characterize the concept of “learning-oriented” path-planning, the underlying configuration space is equipped with certain new classes of exploration inducing distance metrics. These technical foundations will be highlighted through a broad range of aerospace engineering applications with agile maneuvering and robust perception inside dynamic and uncertain environments.

  2. Mr. Naga Chakrapani Pemmaraju, Senior Applications Engineer, MathWorks India Pvt. Ltd.

    Adopting AI in today’s engineering world
    Abstract: AI, or artificial intelligence, is transforming the products we build. It also presents new challenges for those who need to build AI into their systems. Creating an “AI-driven” system requires more than developing intelligent algorithms. It also requires (i) insights from domain experts to generate the tests, models, and scenarios required to build confidence in the overall system, (ii) Implementation details including data preparation, compute-platform selection, modeling and simulation, and automatic code generation, and (iii) integration into the final engineered systemIn this talk , Naga Pemmaraju demonstrates how engineers and scientists are using MATLAB® and Simulink® to successfully design and incorporate AI into the next generation of engineering and medical systems.
  3. Dr. Kallol Roy, CEO and Managing Director, BHAVINI

    A Total Operation & Maintenance (O&M) Management Approach for Plant Systems, Structures & Equipment (SSEs), using System Identification & Bayesian Estimation Framework
    Abstract: The Total O&M Management approach, typically includes, on-line/real-time performance evaluation of process variables, vibration monitoring of roto-dynamic equipment, assessment of stressors for static devices and measurement of loading/strains of structural members & piping. An integrated performance monitoring of all plant SSEs, in real-time, not only helps in improving the availability factor for the entire predicted operating life of the plant, but also helps in subsequent ageing management & residual life assessment of all SSEs and thereby in extending the useful operating plant life, for better economic advantage. Thus, apart from real-time data acquisition, it becomes essential to carry out on-line parametric identification and estimation of unmeasured variables of all plant SSEs and appropriate application of data analytics, decision statistics & distance measures for their performance optimization.For this it becomes necessary to formulate mathematico-statistical models of the governing system dynamics as difference equations or in the state-space framework (using a combination of mechanistic, data-oriented & empirical approaches with additive or multiplicative uncertainties, where the coefficients of ODEs/PDEs and/or the state variables represent specific system parameters) and subsequent identification & estimation of both parameter & states, through a recursive predictor-corrector algorithm. Subsequently, by effective noise separation from the signals/data obtained from dynamic systems/ equipment or through superimposed perturbations on structural members (by use of electro-dynamic shakers), a parametric identification approach for the governing difference equations, are obtained. A Bayesian estimation approach is then applied for estimation of many of the non-measurable parameters or for effectively filtering the noisy parameters (parameters with uncertainty) by application of a Kalman filter (or its variants) or by a Sequential Monte-Carlo Filtering (Particle Filtering) approach. Considering the growth in embedded sensor technology with built-in wireless-sensors or IOT devices, it is now possible to obtain data from various SSEs, in real-time, which can help in both performance assessment/optimization and also effective maintenance planning. Further, the use of such available on-line/real-time data along with off-line simulators, helps in development of digital twins, which serves as an enabler for improving the total O&M maintenance management paradigm.
  4. Mr. Suraj Prakash, Lead Systems Engineer, Honeywell Process Solutions

    Influence of Emerging Technologies on Distributed Control Systems
    Abstract: The talk shall cover briefly the evolution of industrial automation systems and cover the details of the how the emerging technology like virtualization, microservices, cyber, IoT technologies, AI are disrupting the rock steady solutions running for decades in plants. The speaker shall discuss how these technologies shaping up industrial automation solutions . The speaker shall also discuss how the risk averse customers in this domain are receiving this? How would these solution evolve and what more is required to build the confidence in customers to embrace these disruption in the industrial automation space.

Comments are closed.