Manivannan P. V.

A two-degree freedom air fuel ratio controller (Model based feed forward transient plus closed loop Proportional Integral-Derivative (PID) steady state controllers) developed for controlling the air fuel ratio of the charge in a small displacement (125 CC) SI engine is presented. The feed forward controller's airflow and injector models were developed after conducting extensive experiments on the engine modified for the Port Fuel Injection (PFI) operation. A dynamic air fuel ratio model obtained (air fuel ratio changes measured using an UEGO sensor) by injecting the Pseudo Random Binary Signal (PRBS) signal in addition to base line fuel injection pulse, was used for designing the PID controller. Optimal PID gain values were identified using Nelfer-Mead optimization technique. The control algorithms were implemented and optimized using SIMULINK blocks that are run under dSPACE on the MicroAuto box hardware. The optimized control algorithms were ported on the specially designed, in-house built, low cost engine management system (EMS) developed around an 8-bit microcontroller. The spark timing was also controlled simultaneously for knock free operation. The two-degree freedom air fuel ratio controller could maintain the air fuel ratio under steady and transient conditions closely. High thermal efficiency and low HC & NOx emissions were achieved using the developed EMS. At higher speed elevated NOx emission was observed, due to the use of leaner mixture. The improvements are expected to be higher if a suitable smaller injector is used. Copyright © 2012 by ASME.

Using SONAR as the primary range finding sensor has largely been abandoned due to problems such as limited range, large bearing errors and large beam widths. However, SONAR is used conjunction with other sensors such as LIDARs, RADARs and vision sensors for ranging and obstacle avoidance in many autonomous vehicle applications. In this paper, we propose a solution to reduce the range and bearing error significantly, and thus improve the performance of the SONAR. Using the results from the Gaussian Correlation Inequality, we derive probabilistic transformations that can improve the range and bearing measurement of the SONAR, thus reducing the sensor error. We are also presenting simulation study, to place bounds on the types and characteristics of the SONARs within which our model's performance is optimal.

An idle speed engine model has been proposed and applied for the development of an idle speed controller for a 125 cc two wheeler spark ignition engine. The procedure uses the measured Indicated Mean Effective Pressure (IMEP) at different speeds at a constant fuel rate and throttle position obtained by varying the spark timing. At idling conditions, IMEP corresponds to the friction mean effective pressure. A retardation test was conducted to determine the moment of inertia of the engine. Using these data, a model for simulating the idle speed fluctuations, when there are unknown torque disturbances, was developed. This model was successfully applied to the development of a closed loop idle speed controller based on spark timing. The controller was then implemented on a dSPACE Micro Autobox on the actual engine. The Proportional Derivative Integral (PID) controller parameters obtained from the model were found to match fairly well with the experimental values, indicating the usefulness of the developed idle speed model. Finally, the optimized idle speed control algorithm was embedded in and successfully demonstrated with an in-house built, low cost engine management system (EMS) specifically designed for two-wheeler applications. © 2011 The Korean Society of Automotive Engineers and Springer-Verlag Berlin Heidelberg.

This paper presents a novel Simplified Node decomposition and Platoon (SNAP) head selection cluster-based routing algorithm for Vehicular Ad hoc Networks (VANET) communication. The major novelty of the paper is the two proposed algorithms through which the number of platoons and platoon heads (PH) are decided upon for increasing the network lifetime. Algorithm-1 analyze the distribution of nodes and forms a set of possible minimum number of platoons (clusters) having equal number nodes, based on Hierarchical clustering technique. The algorithm-2 partitions the given network into platoons and selects the platoon head (PH) based on coverage range of wireless node. It uses SPSS statistical software tool for platoon decomposition and also to determine the platoon head (PH). The number of iterations required for selection of PH is found to be minimum, irrespective of the number of network nodes. The platoon is controlled by its platoon head (PH) and the data transfer happens locally between the platoon nodes and its head. The platoon head alone communicates with the road side unit (RSU) and avoids multiple data link between the individual nodes of platoon to RSU. The performance of SNAP algorithm has been verified using NS-2 network simulator and found to consume less transmission energy, increased node and network lifetime which in turn increase the overall efficiency of the network.

We report a planar solenoid actuated valveless micropump with multiple inlet-outlet configurations. The self-priming characteristics of the multiple inlet-multiple outlet micropump are studied. The filling dynamics of the micropump chamber during start-up and the effects of fluid viscosity, voltage and frequency on the dynamics are investigated. Numerical simulations for multiple inlet-multiple outlet micropumps are carried out using fluid structure algorithm. With DI water and at 5.0 Vp-p, 20 Hz frequency, the two inlet-two outlet micropump provides a maximum flow rate of 336 μl min-1 and maximum back pressure of 441 Pa. Performance characteristics of the two inlet-two outlet micropump are studied for aqueous fluids of different viscosity. Transport of biological cell lines and diluted blood samples are demonstrated; the flow rate-frequency characteristics are studied. Viability of cells during pumping with multiple inlet multiple outlet configuration is also studied in this work, which shows 100% of cells are viable. Application of the proposed micropump for simultaneous pumping, mixing and distribution of fluids is demonstrated. The proposed integrated, standalone and portable micropump is suitable for drug delivery, lab-on-chip and micro-total-analysis applications.

A cognitive decision making algorithm, which tries to emulate human driver behaviour, for autonomous vehicles is presented in this paper. The decision making process, as well as the integration of the decision making with real time control in different traffic scenarios is studied through simulations using vehicle dynamics model for a car like vehicle. The decision making algorithm was found to perform better, when integrated with the ANFIS Fuzzy Logic longitudinal Control than with conventional PID control. The simulation results with the different traffic scenarios demonstrate the successful integration of the real time control and decision making modules of the Intelligent Vehicle Architecture.

A binary-type exhaust gas oxygen (BEGO) sensor that is usually used to maintain the air–fuel ratio (AFR) at stoichiometric conditions in current spark ignition (SI) engines is significantly lower in cost compared with the universal exhaust gas oxygen sensor. However, it can only switch from a high state (≈0.7 V) to a low state (≈0.1 V) when the mixture goes from richer to leaner than stoichiometric conditions or vice versa. Thus, it cannot indicate the actual AFR. A novel method of estimating the AFR of an SI engine using a BEGO sensor has been demonstrated in this work for leaner than stoichiometric mixtures. Experiments were conducted on a single-cylinder, manifold-injection SI engine. The air–fuel mixture was initially kept at a richer than stoichiometric level while the engine was maintained at constant speed and throttle. The mixture was suddenly changed to a leaner than stoichiometric level. The switching time of the BEGO sensor during this operation was noted. This was repeated for different initial and final AFRs. A relationship between the switching time and change in AFR was obtained for different initial AFRs. A look-up table to determine the AFR was made and used under test conditions. The error is less than 5% in the estimated AFR. The system was also incorporated on a low-cost microcontroller-based engine management system and tested under laboratory conditions. © 2013, SAGE Publications. All rights reserved.

The Prototype Fast Breeder Reactor (PFBR) has eight Steam Generators (SG), each with 547 tubes connecting top and bottom headers. The integrity of these tubes is ascertained using periodic In-Service Inspection (ISI) procedures. Two-Axis Tube Locator Module (TLM) is a robotic manipulator designed to be used to precisely locate all 547 tubes. The difference in the device work plane and the tube sheet plane is mainly due to error from tolerances in fabrication and error due to mounting the device. These errors hinder the accurate positioning of inspection device over the tubes for inspection. In this work, a novel approach to calibrate the TLM is attempted by using motor encoder as sensors, to calibrate the TLM without using any external calibration devices. In this work, a three-point calibration is used to compute transformation matrix using two-axis motor encoder as sensors, for measuring the displaced position of calibration point tubes and using SVD on the correlation matrix formed from the original and transformed points. Even though three independent measurements (x, y, z axis) are required to spatially ascertain the positioning error of the probe with respect to calibration point tube, it proved that planar 2D measurement using readily available servo motor encoder sensors itself is adequate for the calibration of TLM without much loss of accuracies. The degenerated sensing reduces the cost, effort, and time for calibrating the TLM for use in inspection.

Unmanned Aerial Vehicles (UAVs) require advanced path planning and obstacle avoidance algorithms for navigation. In this paper, the Flood Fill Method (FFM) was employed for navigating an autonomous quadcopter in a simulated environment created using MATLAB. The proposed method was compared with the traditionally used Potential Field Method (PFM). For a known terrain, Flood Fill Method works computationally faster for a repetitively same path. These methods were compared in terms of the parameters such as avoiding random obstacles for variable starting and destination positions, the time required to complete the journey, and the optimum path selection.

This paper presents the development of a system for analyzing and adapting the Human Arm motion for virtual reality applications. The proposed system consists of number of Flex sensors, Inertial Measurement Units and Embedded Data Acquisition System, to record the joint angles for deriving the kinematic states (position, velocity and acceleration) of different parts of the human arm. A flexible structure is used for holding the sensors on the human arm at the required position, without hindering the movement. The embedded circuit utilizes a 32-bit Microcontroller to process the data from various sensors and transmits digitized data to the central computer for computing the various kinematics parameters. The system has been tested against standard motion tracking device and is found to perform close to the reference device with an average error of 6%. Such a device can be used to simulate critical operations in medicine and industry and analyze performance during various tasks. © (2014) Trans Tech Publications, Switzerland.