Control Instrumentation Systems

Fuel cells have been considered as an ideal source of energy in the future power generation applications due to its pollution-free nature, noise-free operation and better efficiency. Direct storage of hydrogen in specially designed tanks for automobiles running on fuel cells is not a viable option due to several drawbacks associated with safety and space limitations. To overcome the challenges of direct onboard storage of hydrogen, storing hydrocarbons rich in hydrogen and suitably reforming it to produce hydrogen using several reforming techniques seems to be an acceptable option. Using available gas purification techniques such as palladium membrane-based gas separation, pure hydrogen gas can be extracted from a mixture of other gases and can be fed to the fuel cell for generating power. In this work, a mathematical model of the battery system is analyzed along with a switching controller operating based on an energy management policy. The switching controller switches between battery and fuel cell to ensure a delay-free delivery of the power to the external load. A case study on the dynamic behavior of the integrated system under set point changes in the power demand is analyzed in the presence of a battery backup and a switching controller.

In this paper design of fuzzy logic controller (FLC) on system on chip field programmable gate array (SoC FPGA) for Shell and Tube Heat Exchanger (STHE) is described. STHE is interfaced with Compact Reconfigurable Input Output (CRIO) based FPGA through data acquisition (DAQ) card. For designing controller on chip, National Instruments (NI) CRIO-9101 is reconfigured with FLC code. Interfacing is carried out with NI analog input-output device embedded with NI CRIO-9012 microcontroller and NI CRIO-9101 FPGA module. A graphical program is developed for real-time control of the process plant using CRIO based FPGA. FPGA acts as a standalone processor instead of PC for controlling the STHE. The fuzzy logic controller designed on FPGA is compared with benchmark controller in real time temperature control of STHE. It is demonstrated that the controller proposed outperformed conventional controller.

In this paper, we propose a problem of simultaneous exploration and coverage for a mobile robot, combining the problems of area coverage with exploration and mapping. The primary task here is to completely cover an initially unknown region. Here we combine the advantages of online and off-line coverage path planning algorithms by using the exploration as an aid. The robots perform intermittent exploration during coverage in order to update the map of the environment, which in turn is used to generate the coverage path. We illustrate and demonstrate the problem using the off-line version of Spanning Tree Coverage algorithm with a frontier-based exploration strategy. The simulation results demonstrate that the robot successfully achieves complete and non-repetitive coverage.

Manipulators are widely used in all areas of science and technology. Effective trajectory tracking and quick deflection suppression are the two main aspect of research for a flexible manipulator. The paper reports aperiodic signal like trajectory tracking control for a planar assumed modes modelled two-link flexible manipulator (TLFM). The aperiodic chaotic signal is used as a desired trajectory for the TLFM. Thus, designing of a robust controller for the aperiodic signal tracking control is a challenging task. A backstepping based adaptive SMC technique is designed for the considered problem. In adaptive SMC, the gain of the switching control law is estimated online. The effectiveness of the considered controller is compared to an available backstepping controller. It is found that the designed backstepping based adaptive SMC perform better in terms of smaller tracking time, quick tip deflection suppression and lesser, smoother control efforts. Proposed trajectory strategy is validated on a two-link flexible manipulator in MATLAB simulation environment.

In this paper we propose Geodesic-VPC, a “partition” and “cover” strategy for a multi-robot system using Voronoi partitioning based on geodesic distance metric in the place of the usual Euclidean distance. Each robot is responsible for covering the corresponding geodesic-Voronoi cell using a single-robot coverage strategy. The proposed partitioning scheme ensures that Voronoi cells are contiguous even in the presence of obstacles. We demonstrate that if the single-robot coverage strategy is capable of providing a complete and non-repetitive coverage, then the proposed Geodesic-VPC strategy provides a complete and non-repetitive coverage. We use spanning tree-based coverage algorithm as the underlying single-robot coverage strategy for the purpose of demonstration, though any existing single-robot coverage algorithm can be used.

Objectives of the paper are to (i) develop a new hyperchaotic system having hidden attractors and (ii) to show the applications using the new system in the form of secure communication. New system proposed in the paper has a stable equilibrium, hence considered under the class of the hidden attractors dynamical system. Dynamical characteristics of the novel system is confirmed using some numerical means like phase portrait, Poincaré map and Lyapunov spectrum plot. The applications of the new system are shown by encrypting and decrypting a sinusoidal signal and sound wave. Secure communication is achieved by designing a proportional integral (PI) based sliding mode control (SMC). MATLAB simulation results validate and ensure that the objectives are satisfied.

In this paper we address the problem of area coverage using multiple cooperating robots. One of the main concerns of using multiple robots is of avoiding repetitive coverage apart from complete coverage of the given area. Partitioning the area to be covered into cells and allotting one each cell to each of the robots for coverage is a simple and elegant solution for this problem. However, the spacial partitioning may lead to additional problems leading to either incomplete coverage or coverage overlap near the partition boundary. We propose a manhattan distance based Voronoi partitioning scheme of \(2D\times 2D\) gridded region, where D is the size of the robot footprint. We show that the proposed partitioning scheme completely eliminates coverage gaps and coverage overlap using illustrative results.

The growth properties of ZnO Nanorods was studied on different seed layers by the deposition of Zinc oxide (ZnO) thin film on SiO₂/Si substrate by RF sputtering at two different conditions, i.e., one at room temperature and another at 400 °C substrate temperature. Surface morphology of the seed layer was studied by X-Ray Diffraction and Atomic Force Microscopy (AFM). Low cost hydrothermal method was employed for the growth of ZnO Nanorods on both the seed layers. The structural properties of ZnO nanorods were characterized by Field Emission Scanning Electron Microscope (FESEM). The FESEM images showed the proper alignment and orientation of ZnO nanorods grown on both the seed layers. The I-V measurements were carried out at room temperature under dark light and Ultraviolet (UV) light source. In order to examine the UV detection, MSM (Metal–Semiconductor–Metal) photodetector was fabricated and responsivity was measured for the nanorods grown on both seed layers. The better responsivity and contrast ratio of ZnO nanorods based UV detector was observed in case of 150 nm seed layer deposited at 400 °C.

Modular robots are designed to increase the utilization of robots by modularizing their architecture. We discuss manually reconfigurable manipulators, where a manipulator of desired kinematic configuration is built by assembling the available modules. In the case of a serial-link manipulator with revolute joints, the joint angle is a variable. Out of the remaining three D-H parameters, namely, link-length, link-offset and link-twist, the twist angle influences the workspace the most. This work proposes a conceptual design and fabrication of individual modules which can be assembled to obtain a modular manipulator with desired kinematic configuration in terms of twist angles between any two consecutive joints. We also discuss possible provisions for length adjustment of a link. Designed modules are fabricated using 3D printer. As we focus on manually reconfigurable manipulators, simplicity of individual modules, in terms design, fabrication, and assembly, has been given higher priority, in contrast to similar designs available in the literature.

A unique methodology employing a linear phase finite impulse response (FIR) low pass filter (LPF) was proposed with an attempt to mitigate passband and stopband ripples due to Gibb’s phenomenon. The three regions of the filter response in the frequency domain are approximated using trigonometric functions. The proposed filter model achieved a sharp transition of 2π, fairly flat passband and a stopband attenuation of 40 dB. Our algorithm suppressed the oscillations near the edge of the transition region as well as in the passband region, reducing the Gibb’s phenomenon from the conventional passband ripples from 18% to as low as 2%. Thus a threefold satisfactory performance was achieved in all the three bands namely passband, transition and stopband. Our proposed linear phase FIR LPF was effectively used to filter out power line interference and higher unwanted frequencies from the real time electroencephalogram signals.

Multivariate Statistical Process Control is a projection method of projecting a high- dimensional model space with a number of measured variables to a low-dimensional space. The different methods include Principal Component Analysis, Principal Component Regression, Partial Least Squares Regression, Finite Impulse response, Autoregressive exogenous input and autoregressive moving average, etc. The advantage of using Multivariate Statistical Process Control is that it identifies low-dimensional quality process data while reducing the variability in the process and increasing the product quality. The paper aims to find the low- dimensional information-rich space for soft sensor design using Partial Least Squares-based Multivariate Statistical Process Control technique for the controlling variables in a Wastewater Treatment Plant. The input considered here is the stored data of the actual process variables obtained from the plant. This is carried out using the 14 days data for the three weather conditions, dry rain and storm available from the benchmark model. The performance of the applied method is verified using scatter plot and R-squared.

Most mobile robotic systems draw power from batteries which have a limited power life. Monitoring the status of the battery power on the robot is therefore important for autonomous robotic systems. The wireless system for monitoring lead–acid battery of obstacle avoidance robot has been developed. The system employs sensors, microcontroller ATMEGA 328, and Bluetooth module. The parameters such as voltage and current of battery are checked by the battery monitoring system during discharging of the battery. The varying discharging voltage and current values of the battery are detected using voltage divider circuit and Hall Effect current sensor, respectively. Battery voltage and current data will be transferred in real time to the microcontroller, and then it will be transmitted to display device using Bluetooth communication. In this work, monitored battery data will be displayed on Microsoft Excel of laptop Personal Computer (PC) with Parallax Data Acquisition tool (PLX-DAQ). The real-time data of voltage and current will be indicated by the system in the tabulated form along with graphical display.

One of the deadliest diseases in human beings is the glaucoma, which is the second largest disease in the world, which leads to the loss of vision in the human eye, thus making the life of human miserable and the whole world would be dark without vision. Recently, (DL) Deep Learning is playing a lot of important role in the image processing applications. This DL can be clubbed with CNNs (Convolution Artificial Neural Networks) along with a hardware Raspberry Pi and the hybrid combination of the threesome could be used for the automated detection of the glaucomatic case in the disease-affected human beings in the eyes. In this write-up, the previously mentioned hybrid threesome is being used and developed for the glaucoma detection. The DL frameworks (CNN + ANN + MATLAB) can be used as a hierarchical representation of the fundus images to distinguish b/w glaucoma and non-glaucomatic images for the disease detections. The model is trained with standard datasets available on the net. The VGG19 architecture is used with transfer learning to achieve high accuracy. A graphical user interface is used to diagnose the condition of test images and give a graphical analysis of the patients. The entire program is run on a Raspberry Pi 3B with a 5” LCD touch screen as a stand-alone device with the power input.