New user registration
Event organizer registration
Login via your institution account
Login via your organization account
Login via your publisher account
Login using your new user credentials
Forgot username or password?
Subscribe to SciArchives
Indian Journal of Automation and Artificial Intelligence
Publisher:Indian society for Education and Environment
Editor in chief:Prof.Natarajan Gajendran
Your selection(s) could not be saved due to an internal error. Please try again.
Browse by Volumes
Data Fusion Of Towed Array And Hull Mounted Array Measurements For Passive Acoustic And Electromagnetic Underwater Localisation And Classification
Author:A. Jawahar, Ch.Rajya Lakshmi.
Volume: 3 | Issue no: 1-2016 | Pagination: 1-5
Background/Objectives: In our earlier work, data fusion with specific application to underwater tracking environment was explored. The target can be tracked using array bearings, while it is moving with constant velocity and maneuvering occasionally. Methods/Statistical analysis: In this paper, it is shown that if data fusion is carried out using the bearing measurements available from Towed Array (TA) along with hull mounted array's bearings, then tracking of a continuously moving target can be carried out easily. Findings: This algorithm is independent of ownship maneuver for the observability of the process. Song and Speyer's modified gain algorithms are utilized with some modifications for estimation. Application/Improvements: Monte Carlo simulation is performed and results are shown for various typical geometries which revealed that this algorithm is useful naval based applications. Keywords: estimation, sonar, tracking, data fusion, Kalman filter, observability, maneuver.
Feasible Course Trajectories For Undersea Sonar Target Tracking Systems
Volume: 3 | Issue no: 1-2016 | Pagination: 1-5
Background/Objectives: In underwater sonar environment, the target motion parameters can be obtained only when observer maneuvers in some particular manner is satisfying certain requirements. Methods/Statistical analysis: The algorithm is evaluated using Line of sight measurements which are obtained from intercept radar. Though the recommended maneuver may not be optimum, observability is ensured. Findings: Recursive Maximum Likelihood Estimator with initial estimation from Recursive Pseudo Linear Estimator is used to evaluate the process. Application/Improvements: For the purpose of analysis, the proposed observer maneuver is used for a typical scenario at low, medium and high target angles. Convergence time and the accuracy of the solution in Monte-Carlo simulation are presented in detail.
Rough Set Theory Approach For Attribute Reduction
Author:Lukshmi R.A, Geetha P.V , Venkatesan P
Volume: 1 | Issue no: 3-2013 | Pagination: 70-80
Knowledge Discovery from databases is practically important in many fields , including the field of medicine. Many methods are being developed for knowledge discovery and due to the availability of enormous amount of data, extraction of knowledge from database has become a challenging task. Researchers have proved methods, among which Rough Set Theory is an effective tool for knowledge discovery. In this paper, Rough Set Theory and its basic ideas are reviewed and applied to identify symptoms for diagnosing diabetes. This study also presents methods for extension to high dimensional data.in the medical domain.
Investigating Anns And Applications
Author:Ahmad Waqas, Abdul Rehman Gilal, Zeeshan Bhatti, Abdul Waheed Mahessar
Volume: 1 | Issue no: 2-2013 | Pagination: 65-69
Artificial Neural Network (ANN) has emerged with advancement of Information and Communication technology and biological sciences during last decades. The aim is to utilize technology and construct machines that will work like brain of humans. The internal architectural requirements of such a machine is to have huge simultaneous memory and storage in consistent with intensive processing power to cater the ambiguous information and behave like human brain. ANN has broad range of applications in today's business and IT industry. This paper aims to investigate the working of ANN and its applications in real environment.
Expert System For Agriculture Extension
Author:Sujai Das, Laxmikanta Nayak
Volume: 1 | Issue no: 2-2013 | Pagination: 62-64
This paper is meant to provide required information and expert advice to the farmers and extension workers according to their needs & available resources. For an example, on the basis of symptoms supplied by the farmer, diseases affecting the crop can be detected and those practices that should be adopted according to the geographical locations or climate for a better yield. Thus, the work is to categorize agriculture in sub-areas & collect relevant information for these areas and feed into database. Second to make decision rules to process the information. Finally to design & develop the web based expert system for extension people. This expert system is expected to provide required information to the farmers and extension workers to take decisions in the agricultural enterprise.
Electromotive Force Can Be Induced By Static Magnets In A Stationary Wire
Volume: 1 | Issue no: 2-2013 | Pagination: 57-61
Electromotive force is the energy per unit charge that is created during the separation of positive and negative charges. As Predicted by Faraday, the Electromotive force cannot be induced by a static Magnetic field. Here I present evidence that electromotive force can be induced by a static magnetic field in stationary wire .The magnets exists in dipoles. The constant magnetic flux from static magnets can induce Electromotive force in a stationary wire. This Electromotive force induced is very minimal which is in micro volts. The force of attraction between the opposite poles of a magnet can move the electrons present in the insulated copper wire. Since the magnetic flux exerted by the magnets is constant, the Electromotive force induced is also constant with respect to time.
Efficient Roi Segmentation Of Digital Mammogram Images Using Otsu'S N Thresholding Method
Author:S.Deepa, Dr.V.Subbiah Bharathi
Volume: 1 | Issue no: 2-2013 | Pagination: 51-56
Segmentation of the Region of Interest (ROI) is the first and crucial step in the analysis of digital mammogram images since the success of any Computer Aided Diagnostic (CADx) system depends greatly on the accuracy of the segmentation of the ROI from the mammogram images. Finding an accurate, robust and efficient ROI segmentation technique still remains a challenge in digital mammography. In this paper we have proposed an efficient Otsu's N thresholding method for segmenting regions of interest from the mammogram images. Digital Mammograms are taken from the mini MIAS (Mammographic Image Analysis Society) database for the purpose of experimentation and the results obtained are scaled to full color. Results show that the proposed method is efficient and is in concurrence with the ground truth table available in the database.
Ascertaining Cognition Abilities Of 1st Generation Cognition Robot Using Bayesian Models
Volume: 1 | Issue no: 2-2013 | Pagination: 44-50
There are number of challenges involved in programming cognition into a robot, e.g. getting the robot to learn about its architecture, being aware of the things around it, acquiring knowledge by itself, and most importantly carrying out trivial tasks like responding to flash lights or running away from the fire are the kinds of activities the robots should be able to do on its own. Given that a large set of variables involved in performing trivial tasks are presented to it, could it derive the relationships between them Using Bayesian or in other words the belief networks a model was developed to ascertain the level of cognition skills acquired. The 1st generation cognition robot based upon a popular atmega32 microcontroller was designed, as a platform to carry out a number of artificial intelligence experiments. Experiments were aimed at relating, mere three sensors with those of stimuli and drawing up a Bayesian network with relevant weights. By repetitive subjection to stimuli, the robot was able to build the network as desired. Further, to ascertain the cognition abilities, 20 relations that formed a branch in the network, were queried multiple of times to find if they conform to the correct response. Data showed that, probability of occurrences of a particular branch being true has a regression fit of 0.73 with the desired response, suggesting that indeed the robot has acquired certain level of cognition.
Fuzzy Based Routing In Wireless Mesh Network
Author:J. Arun Pandian , C. P. Indumathi
Volume: 2 | Issue no: 1-2015 | Pagination: 81-94
Background: Wireless Mesh Network (WMN) is a continuously self-configuring, infrastructure-less network. A fundamental issue arising in Wireless Mesh Network is the selection of the optimal path between any two nodes. A route discovery attempt can possibly result in several paths being uncovered for a single destination. Methods: In this paper, we have proposed trust-based geographical routing algorithms to tackle the cyber-attacks in Smart Grid, which are inspired from the existing Ambient Trust Sensor Routing (ATSR) algorithm. The Fuzzy-based Energy Aware Trust Geographical Routing in Wireless mesh network (FEATGRWMN) is proposed by using the fuzzy logic approach. It can synthesize the trust, energy, and distance metrics to calculate one final score as the routing metric for determining the best route between source and destination nodes. Findings: The extensive simulation studies have confirmed that our new FEATGRWMN algorithm is more timesensitive to detect then avoid the recent malicious attacks than the existing ATSR algorithm. Application: The FEATGRWMN algorithm is able to achieve better routing performance in different network scenarios by solving the weight factor issue.
Comparison Of Decision Tree And Svm Methods In Classification Of Researcher'S Cognitive Styles In Academic Environment
Author:Z. Nematzadeh Balagatabi, H. Nematzadeh Balagatabi
Volume: 1 | Issue no: 1-2013 | Pagination: 31-43
Recently, by development of internet, it is user's right to achieve the best answer based on what they demand. Also, classification is the task which is essential in data mining. Nowadays, there are many classification techniques to eliminate the classification problems such as Decision tree, SVM, Genetic Algorithm, Bayesian and others. In this paper, the researchers are classified to "Expert" and "Novice" based on cognitive style factors to have the best practicable answers. Academic environment has been chosen as a domain of this research. An important aim of this research is to classify the researchers based on Decision tree and Support Vector Machine techniques and finally according to the highest accuracy, choose the best technique to help the researchers to have the best answer based on their request in digital libraries. Background/Objectives: The main concepts of cognitive styles and specifications of Decision tree, and SVM methods are investigated. The implementation for classification of Decision tree and SVM methods are developed, finally, the classification based on the accuracy of the results are compared. Methods/Statistical analysis: Two methods of classification are used in this paper which are Decision tree and Support vector machine. There are various methods in Decision tree, but only 6 of them are used here which includes J48, LMT, RandomForest, REPtree and DecisionStump. Also, the experiment in SVM was based on 10-fold cross validation. To strengthen the analysis in SVM two experiments are done. The new experiment is based on 5-fold cross validation. Results: Based on the achieved values, if these two methods compare with accuracy and average accuracy values, SVM is the best method in comparison with Decision tree. Moreover, it can be concluded that, SVM can classify more precisely than Decision tree, because it categorizes using separating hyperlanes and margins. However, Decision tree does not use hyperlane, so may have some errors in classification. Conclusion/Application: Based on the SVM method, researchers can be classified to "Expert" and "Novice" based on cognitive style factors in order to have as best as possible answers. So, researchers will have the best feedback based on their demands in the digital libraries
Contact and support
Terms and conditions
Frequently asked questions
Copyright © 2015
All rights reserved. SciArchives.