Seshadhri Srinivasan

• PhD
• ABB

We propose TRAINER, a TRustworthy Artificial Intelligence-based iNsulin recommendER for elderly individuals with type 2 diabetes, ensuring reliability and trust in insulin dosage recommendations. TRAINER exemplifies this trustworthiness and addresses such concerns by offering reliable insulin recommendations supported by clinical evidence.

Managing Elderly type 2 diabetes (E-T2D) is challenging due to geriatric conditions (e.g., co-morbidity, multiple drug intake, etc.), and personalization becomes paramount for precision medicine. This paper presents a human digital twin (HDT) framework to manage E-T2D that exploits various patient-specific data and builds a suite of models exploiti...

Industrial networked control system performance can be compromised through introduction of intended attacks by intruders. Several ways have been reported in literature to introduce attacks like denial-of-service, packet dropout, and delays etc. in Industrial networked control system. Attacks may be designed based on the system knowledge gathered th...

OBJECTIVE To derive macronutrient recommendations for remission and prevention of type 2 diabetes (T2D) in Asian Indians using a data-driven optimization approach. RESEARCH DESIGN AND METHODS Dietary, behavioral, and demographic assessments were performed on 18,090 adults participating in the nationally representative, population-based Indian Coun...

Emerging development in networked control systems has been observed by integrating sensors, controllers, and actuators. The networked control systems (NCSs) are used in various fields due to ease of installation, reduced maintenance, system wiring, and diagnosis. However, the real-time communications via a network of NCS drive it towards vulnerable...

An artificial pancreases (AP) is a device for managing diabetes through automated insulin infusion. The control algorithm is the heart of the AP that computes the insulin infusion based on blood glucose measurements. In this article, we investigate the role of multimedia data to enable the advanced control techniques that could personalize AP in el...

A networked control system (NCS) is sensitive to the different types of attacks and it is essential to secure and stabilize it. In this manuscript, to represent the impact on the stability of the control system, an industrial NCS is considered. The process noise and measurement noise alongside certain attacks are assumed to be affecting the perform...

We investigate the role of explainable Artificial Intelligence (XAI) for building trust in data-driven fault detection and diagnosis (FDD). We examine use cases for XAI-FDD on a building in Singapore that has six chillers.

Keshary, ShivomDharmaraj, GaneshaperumalBalasubramanian, SubathraSrinivasan, SeshadhriFrom the last decade, Internet of Things (IoT) brought tremendous changes for industries and health care. It provides the smart digital platform for remote healthcare monitoring. With the help of IoT technology, patients can monitor and record their vital sign par...

This paper presents an eXplainable Fault Detection Systems (XFDS) for incipient faults in PV panels. The XFDS is realizable on simple edge devices and has four main components: (i) irradiance-based three diode model (IB3DM), (ii) data-based fault classifier, (iii) eXplainable Artificial Intelligence (XAI) application, and (iv) edge node implementin...

This paper presents a computationally efficient novel heuristic approach for solving the combined heat and power economic dispatch (CHP-ED) problem in residential buildings considering component interconnections. The proposed solution is meant as a substitute for the cutting-edge approaches, such as model predictive control, where the problem is a...

This article studies a scalable control method for multizone heating, ventilation, and air-conditioning (HVAC) systems to optimize the energy cost for maintaining thermal comfort (TC) and indoor air quality (IAQ) (represented by CO₂) simultaneously. This problem is computationally challenging due to the complex system dynamics, various spatial and...

Industry 4.0 is expected to revolutionize the way industries are automated currently. Conventional Supervisory Control and Data Acquisition (SCADA) systems are connected to various remote terminal units that are geographically and functionally separated in an industry. This makes SCADA system vulnerable to attacks and the real-time timing requireme...

Contribution: This article explores how the Industrial Internet of Things (IIoT) could be leveraged to enhance the teaching/learning experience of advanced control techniques [e.g., model-predictive control (MPC)] for complex systems (nonlinear and multivariable) for undergraduate students. Background: The IIoTs' features, such as ubiquitous sensin...

This paper presents an eXplainable Fault Detection and Diagnosis System (XFDDS) for incipient faults in PV panels. The XFDDS is a hybrid approach that combines the model-based and data-driven framework. Model-based FDD for PV panels lacks high fidelity models at low irradiance conditions for detecting incipient faults. To overcome this, a novel irr...

This study presents an extremum seeking-proportional–integral and derivative (ES-PID) controller design for brushless direct current motors and its implementation in electric vehicles. The ES-PID controller aims to simultaneously maintain a speed set-point and reduce torque ripples in the presence of load-torque disturbances. The proposed ES-PID co...

Formulation of cost function and selection of optimization method are the major aspects of optimization based SLAM, which has been addressed in this paper. Two cost functions namely (i) Map Oblique Error and (ii) Map Spread Error have been formulated for correcting the reconstructed map in SLAM. They are based on the structural regularities of the...

This paper presents an approach to recursively estimate the simplest linear model that approximates the time-varying local behaviors from imperfect (noisy and incomplete) measurements in the internet of things (IoT) based distributed decision-making problems. We first show that the problem of finding the lowest order model for a multi-input single-...

Diabetes technology (DT) has accomplished tremendous progress in the past decades, aiming to convert these technologies as viable treatment options for the benefit of patients with diabetes (PWD). Despite the advances, PWD face multiple challenges with the efficient management of type 1 diabetes. Most of the promising and innovative technological d...

This investigation studies a scalable control method for multi-zone heating, ventilation and air-conditioning (HVAC) systems with the objective to reduce energy cost while satisfying thermal comfort and indoor air quality (IAQ) (represented by CO2) simultaneously. This problem is challenging as we need to cope with the complex system behaviours, va...

Space cooling in buildings is influenced by thermal dynamics, which in turn is affected by ambient conditions, solar radiation, occupancy, stray heating and various other disturbances that are time-varying and nonlinear. This investigation presents an adaptive disturbance observer for estimating the thermal states of the building depending on the d...

This paper presents a novel solution technique for scheduling multi-energy system (MES) in a commercial urban building to perform price-based demand response and reduce energy costs. The MES scheduling problem is formulated as a mixed integer nonlinear program (MINLP), a non-convex NPhard problem with uncertainties due to renewable generation and d...

Maintaining both indoor air quality (IAQ) and thermal comfort in buildings along with optimized energy consumption is a challenging problem. This investigation presents a novel design for hybrid ventilation system enabled by predictive control and soft-sensors to achieve both IAQ and thermal comfort by combining predictive control with demand contr...

Building intelligence in autonomous robots to classify heterogeneous terrains on-the-move is a challenging task, but a pivotal feature required for accomplishing safety critical missions. This paper proposes an adaptive neuro-fuzzy inference system for online terrain classification in the wheeled mobile robot using the steady-state behaviour of rob...

This investigation presents a smart artificial pancreas (AP) for treating Type 1 Diabetes Mellitus (T1DM) in elderly which simultaneously automates the insulin administration but also diet recommender system using implicit carbohydrate (CHO) measurements. Three main components of the AP are: (i) long-term model of physiological dynamics, (ii) model...

This investigation presents an Industrial Internet of Things (IIoT) architecture and a Model-Based Engineering (MBE) approach for design, verification, and auto-code generation of control applications in process industries. The IIoT architecture describes the hardware components, communication modules, and software. It emerges as a major enabler fo...

In precision agriculture, fusing low-resolution Remote Sensing (RS) information with proximal sensors provided by Wireless Sensor Networks (WSN) can help increase productivity, reduce cost, and optimize resources. However, sensor cost and range emerge as a major concern. Consequently, optimal sensor placement has emerged as key. Here we examine it...

Early detection of incipient faults is of vital importance to reducing maintenance costs, saving energy, and enhancing occupant comfort in buildings. Popular supervised learning models such as deep neural networks are considered promising due to their ability to directly learn from labeled fault data; however, it is known that the performance of su...

Scalability of control algorithms used for savings energy in commercial building Heating, Ventilation and Air-Conditioning (HVAC) system and their implementation on low cost resource constrained hardware is a challenging problem. This paper presents the Internet of Things (IoT) prototype which implements a smart and scalable control approach called...

This investigation presents an optimisation-based design of fractional order proportional integral (FO-PI) controller for web transport systems used in paper industries. The objective of the optimisation algorithm is to reduce the integral absolute error of the closed loop web transport systems considering the underlying physical and operating cons...

With the advent of the Internet of Things (IoT), Cloud-based Cyber-Physical Systems (C2PS) are becoming more prevalent in manufacturing systems. Verifying the security of C2PS is a challenging task due to the inclusion of newer entities such as mobile phones and IoT devices. Therefore, scalable methods for verifying security properties of manufactu...

This investigation aims to study different adaptive fuzzy inference algorithms capable of real-time sequential learning and prediction of time-series data. A brief qualitative description of these algorithms namely meta-cognitive fuzzy inference system (McFIS), sequential adaptive fuzzy inference system (SAFIS) and evolving Takagi-Sugeno (ETS) mode...

Crude oil pricing models are frequently studied in energy economics through classical linear regression models subject to various limitations (e.g., normality, stationarity) and diagnostic evidence (e.g., information criterion Occams razor principle). In contrast to conventional practices, sparse identification approach makes a breakthrough in econ...

The paper suggests a simple energy saving controller for heating, ventilation, and air-conditioning (HVAC) systems that combines information on occupancy and weather with predictive control to save energy in buildings. The controller uses a pulse width modulation strategy and turns on/off the HVAC system based on the optimal decisions of the model...

This paper presents an adaptive linear quadratic regulator (LQR) for networked control systems that varies its gains based on the estimates of the time-varying network delays. A sequential learning algorithm for minimum radial basis function neural network, called the minimum resource allocation network (MRAN) is used to estimate the time delays on...

This investigation proposes a CPES architecture and model for engineering energy management application for smart grids. In particular, the investigation considers the implementation of the power systems state estimator (PSSE). The CPES architecture has three layers: physical, monitoring and applications. The physical layer consists of the grid and...

Time-triggering and distributed nature of the grid are emerging as the major challenge in managing energy in distribution grids. This investigation presents an event triggered distributed optimal power flow (OPF) algorithm for energy grids. To generate the event triggers, we use the epidemic algorithm. The buses are classified into three: infected,...

Smart Grid provides a flexible and powerful framework to integrate Distributed Energy Resources (DER). DER consisting of energy sources (such as renewable, conventional, storage) and loads. Based on application, there can be different modes of operating. It can operate in parallel with, or independently from, the main grid. Because of social, econo...

Wheel slip affects the accuracy of dead-reckoning based localization techniques as they introduce measurement errors in odometers. This investigation presents a new slip compensation scheme that uses neuro-fuzzy technique for self-calibration of odometer. The proposed self calibration procedure can be executed in robot navigating environment rather...

Two Major challenges in securing reliable Optimal Power Flow (OPF) operations are: (i) fluctuations induced due to renewable generators and energy demand, and (ii) interaction and interoperability among the different entities. Addressing these issues requires handling both physical (e.g., power flows) and cyber aspects (computing and communication)...

This investigation proposes an energy management system for large multizone commercial buildings that combines distributed optimization with the adaptive learning. While the distributed optimization provides scalability and models the fresh-air infusion as ventilation constraints, the learning algorithm simultaneously captures the influences of occ...

Machine type communication (MTC) systems are a new paradigm in communication systems where machines talk to each other rather than humans. It is expected that more than twenty billion smart devices are deployed around the globe by 2020. The machines talk to each other and communicate with cloud based MTC servers to monitor and control everything ar...

This paper proposes a stochastic optimal controller for networked control systems (NCS) with unknown dynamics and medium access constraints. The medium access constraint of NCS is modelled as a Markov Decision Process (MDP) that switches modes depending the channel access to the actuators. We then show that using the MDP assumption, the NCS with me...

This investigation presents a fault diagnosis methodology for detecting sensor faults in cement industries pyro processing section. It works in three steps: (a) modelling, (b) analysis, and (c) validation. In the modelling, the actual data from the cement pyro processing is used to do a correlation analysis between output and input variables. The s...

Operating heating power plant (DHPP) with fluctuating load is a complex problem. Thermal energy storage (TES), flexible loads, and operating constraints compound this complexity further. This investigation focuses on the design of a model predictive controller (MPC) that reduces the operating and maintenance cost in a DHPP, considering TES and flex...

This paper proposes a stochastic optimal controller for networked control systems (NCS) with unknown dynamics and medium access constraints. The medium access constraint of NCS is modelled as a Markov Decision Process (MDP) that switches modes depending the channel access to the actuators. We then show that using the MDP assumption, the NCS with me...

Energy management in electric grids with multiple energy sources, generators, storage devices, and interacting loads along with their complex behaviours requires grid wide control. Communication infrastructure that aggregates information from heterogeneous devices in the electric grid making the applications completely independent of physical conne...

Cement grinding in ball-mill consumes majority of the energy in cement industry. Current models in literature capturing the material flow are not suitable for designing predictive controllers for energy savings. This investigation proposes two data-driven modelling approaches for cement grinding process that relate material flow and energy. Data ob...

In this paper, we study the problem of feedback mechanism design for packet retransmission in lossy wireless networks where nodes employ intra-session random linear network coding (RLNC). Using intra-session RLNC, intermediate nodes transmit coded packets by combining the packets of various flows until they receive acknowledgements. A delayed ackno...

Autonomous vehicle navigation in an unfamiliar environment is a challenging task. This investigation presents two algorithms for autonomous vehicle navigation in unknown environments with obstacles. Basic building blocks of the path-planning algorithms are the behaviour-based controller and nonlinear state estimator. Measurements from the sensor mo...

This paper presents a direct load control based demand side management (DSM) algorithm that performs peak shaving considering time-varying renewable generation, and thermal comfort of the buildings. The demand side operator of the microgrid (MG) uses the DSM algorithm for peak-shaving, and reducing the energy costs. The DSM controller has a hierarc...

Sensor fusion based localization techniques often need accurate estimate of the fast and uncertain scene change in environment. To determine the scene change from two consecutive LIDAR scans, this paper proposes a novel technique called 'keep zero as zero' polar correlation. As it name implies any zero in the scan data is kept isolated from scene c...

Autonomous vehicles (AVs) manoeuvring in unknown environment require path-planning algorithms that are safe, yet optimal to circumvent dynamic obstacles with minimum fuel-cost. This investigation presents an autonomous vehicle path-planning (AVPP) controller that uses mixed integer linear programming to decide the blending and switching actions amo...

This investigation presents a personalized energy management system (PEMS) for heating, ventilation and air-conditioning (HVAC) systems in residential buildings based on economic model predictive control (EMPC) integrated with occupancy and occupant behavior. The major building blocks of the PEMS are: weather forecasting tool, occupancy predictor,...

Wheel slip compensation is vital for building accurate and reliable dead reckoning based robot localization and mapping algorithms. This investigation presents stochastic slip compensation scheme for robot localization and mapping. Main idea of the slip compensation technique is to use wheel-slip data obtained from experiments to model the variatio...

In web transport systems (WTS), parameter variations in transported material affects the product quality and integrity of the processed material. This investigation presents an adaptive model predictive controller for WTS in process industries considering the variations of the web radius with respect to time. The proposed controller uses a radius a...

Studying cyber-physical system (CPS) for a given network protocol and processor schedules is a challenging task. This investigation illustrates the role of TrueTime a MATLAB package for simulating CPS encapsulating information on processor schedules, and network protocols. Properties of CPS such as temporal behaviors, performance and stability can...

In this paper, Tuning of centralized PID controller using Dynamic State Transition algorithms (DSTA) is presented. DSTA is an upgraded version of STA. Centralized PID controller design problem is framed as an optimization problem by minimizing integral of the Absolute Error (IAE). Comparisons are also made with CMAES and RGA based approach. DSTA is...

Path-planning algorithms that can circumvent moving obstacles are required for realizing reliable autonomous vehicles. This investigation presents a path-planning algorithm for autonomous vehicle (AV) that uses infrastructure-to-vehicle (I2V) communication, Extended Kalman Filter (EKF) and behaviour based hybrid controllers for circumventing moving...