APPLICATION OF THE TAKAGI-SUGENO FUZZY CONTROLLER FOR SOLVING THE ROBOTS' INVERSE KINEMATICS PROBLEM UDC

Size: px
Start display at page:

Download "APPLICATION OF THE TAKAGI-SUGENO FUZZY CONTROLLER FOR SOLVING THE ROBOTS' INVERSE KINEMATICS PROBLEM UDC"

Transcription

1 FACTA UNIVERSITATIS Series: Mechanics, Automatic Control and Robotics Vol.3, N o 5, 3, APPLICATION OF THE TAKAGI-SUGENO FUZZY CONTROLLER FOR SOLVING THE ROBOTS' INVERSE KINEMATICS PROBLEM UDC Vesna Ranković, Ilija Nikolić Faculty of Mechanical Engineering, Sestre Janji} 6, 34 Kragujevac, Serbia and Montenegro, inikolic@tt.yu Abstract. For maing from external to internal coordinates the Takagi-Sugeno controller is used, imlemented within the framework of adative network, which is similar, by its architecture, to RBF neural network, so it is also called the neuro-fuy system. During the learning rocess, the arameters of the membershi functions of the rimary fuy sets and arameters of the consequences of the Takagi-Sugeno controller were adated. The results are resented for the straight-line trajectory tracking, with the constant velocity of the grier and with the triangular velocity rofile.. INTRODUCTION Deficiencies of analytical and numerical methods of solving the inverse kinematics (IK) roblem led to searching of the new aroaches to maing from external to internal coordinates. A large number of aers are devoted to alication of the non-recurrent (or feed forward) neural networks for solving the IK. In [6] is used the two-layered neural network (in literature also known as Radial Basis Function RBF neural network). The dynamic rocedure was alied for learning the network. Neurons in the invisible layer have Gaussian activation functions, while the outut layer consists of a single neuron whose activation threshold is being adated. As it is shown, the main deficiency of the RBF networks, trained by learning data sets, which enhance the art of the working sace, lies in fact that their hidden layers contain large number of neurons. The RBF neural networks cannot realie maing from external to internal coordinates in real time for high grier velocities. Fuy logic was first alied for solving the IK of the lanar four-segment redundant robot in aer [4]. This aer was used as a basis for develoing the fuy logical controller (FLC), resented in aer [5], where the FLC is alied for maing from the external to internal coordinates for lanar two-segment robot and for robots of the R T T y and R R y R y Received Aril 3,

2 4 V. RANKOVIĆ, I. NIKOLIĆ minimal configuration. The biggest deficiency of such an FLC structure is the long comutational time, which is the consequence of the large number of stes during the trajectory generation. In addition, in design of logical fuy controller, esecially roblematic is defining of the rule base and selection of the membershi functions of the rimary fuy sets. Numerous researchers have tried to automatie the modeling rocess of the fuy controller. Because of such investigations the adative fuy controllers aeared. Adating mechanism decides which changes should be erformed in the fuy controller, so it would obtain the desired outut, namely it would minimie the error. This mechanism is similar to the training mechanism in neural networks. Different adative systems were develoed. The most frequently used adating mechanism erforms the modification in the hase of controlling rules, changes of the weighting functions of the joined fuy rules to controlling rules in the rule base, modification of the rimary fuy sets or the selection of the defuyfication methods. Based on available literature, it can be concluded that the different training methods were alied most frequently to the Takagi-Sugeno controller. In this work is used the Takagi-Sugeno controller imlemented within the framework of the adative network for solving the IK roblem. In the second section of the aer is resented the basic structure of the controller, and is briefly described the rocedure used for training. In the third section are resented results of maing from external to internal coordinates for the three-segment robot of the R R y R y minimal configuration. The fourth section contains the concluding remarks.. TAKAGI-SUGENO FUZZY CONTROLLER AND THE LEARNING ALGORITHM The Takagi-Sugeno controller with M inuts and one outut is shown in Figure. The fuy controller has M+ linguistic variables: M inut ones and one outut variable. Linguistic variables x i are A i, A i,, A ni, M. For the system with M inuts and one outut, the set of linguistic rules is defined in the form: R : if x is A and x is A and x M is A M, then f = x + x + + M + c R : if x is A and x is A and x M is A M, then f = x + x + + M + c R : if x is A and x is A and x M is A km, then f k = k x + k x + + km + c k R : if x is A n and x is A n and x M is A nm, then f = x + x + + M + c The number of linguistic rules is = n M. The outut from Takagi-Sugeno controller from Figure is: where: y = u f () u i = u i i u = µ ( x )* µ ( x )...* µ ( x ) M u i i M u = µ ( x)* µ ( x)...* µ M( xm)... (3) ()

3 Alication of the Takagi-Sugeno Fuy Controller for Solving the Robots' Inverse Kinematics Problem 4 * denotes certain T-norm. uk = µ ( x )* µ ( x)...* µ M( xm)... u = µ ( x )* µ ( x )...* µ ( x ) n n nm M x x x M A a,b,c u u A a,b,c µ µ µ M x x x M uf x, x,... x M A M,c M a M,b M u u uf xx,,... x M x x x M A a,b,c µ A µ µ M u k u k k a,b,c u f x, x,... x k k M x x A M a M,b M,c M y x M x x x M A n,c n a n,b n A n µ n µ n u u u f x, x,... x M a n,b n,c n µ nm A nm a nm, b nm, c nm Fig.. Takagi-Sugeno controller with M inuts and one outut. If the membershi functions are taken in the Gaussian form then:

4 4 V. RANKOVIĆ, I. NIKOLIĆ µ ij =,,,, b ij i = n j = M xj a ij + c ij Consequences functions of the fuy rules are of the form:. (4) M f = x + c (5) i ij j i j = Substituting () into () the controller outut is obtained as: y = ui fi i u = i or, substituting (5) into (6), the outut of thetakagi-sugeno controler is: (6) M y = u x + c. (7) i ij j i i j u = = i In [], for adating arameters (a, a,, a M, b, b,, b M, c, c,, c M ) is used the iterative rocedure, which is based on method of decreasing gradients. The arameters of the f i functions (,,, M, c,,,, M, c,, k, k,, km, c k,,,, M, c ) are being adated by the least squares method. The total number of arameters for adating is: 3nM + n M (M+). The learning method requires set of data for training P = {,,, r ). Each element of the set, k = (x k, y k ) is defined by the inut vector x k = (x k x k x Mk ) and desired resonse y k. If the FLC arameters correction is erformed after resenting all the samles, then the error sum of squares is: or: where: r k k = ε= ε (8) r ( yk yk ) (9) k = ε= y = u f k ik ik uik M ik ij jk i j = () f = x + c () u =µ ( x )* µ ( x )...* µ ( x ) k k k M Mk uk =µ ( x k)* µ ( xk)...* µ M( xmk)... ()

5 Alication of the Takagi-Sugeno Fuy Controller for Solving the Robots' Inverse Kinematics Problem 43 ukk =µ ( x k )* µ ( xk )...* µ M ( xmk )... uk =µ n( xk )* µ n( xk )...* µ nm ( xmk ) Minimiation of the function given with (8) can be realied through the following iterative rocedure for arameters adating: ε aq ( l+ ) = aq ( l) ηa aq ε bq ( l+ ) = bq ( l) ηb bq (3) ε cq ( l+ ) = cq ( l) η cq From equations (3) one obtains: r yk aq ( l+ ) = aq ( l) ηa ( yk yk ) k = aq (4) where: yk uik u ik = fik yk a q aq a q u where: where: ik r b ( l+ ) = b ( l) η ( y y ) y k q q b k k (5) k = bq y u u b b b k ik ik = fik yk q q q u ik r c ( l+ ) = c ( l) ( y y ) y k q q η c k k (6) k = cq y u u c c c k ik ik = fik yk q q q u ik Parameters of the f i functions are being determined by the least square method. From equations () and (5) one obtains that the resonse of the fuy controller to the k-th element is: yk = uk( xk + xk +... MxMk + c ) + uk( x k + xk +... MxMk + c ) +... (7) u ( x + x +... x + c ) k k k M Mk The aim of learning is that the real resonse is equal to the desired one. If the adating is done after resenting of all samles, then:.

6 44 V. RANKOVIĆ, I. NIKOLIĆ ux ux! u xm u ux ux! uxm u! ux ux! uxm u ux ux! uxm u! " "! " " " "! " "! urx r u rxr! u rxmr u r urx r urxr! urxmr ur! " M c u x u x! u xm u y ux ux! uxm u y " = " "! " " M " ur x r ur xr! ur xm ur c yr " " M c Equation (8) is resented in the form: UP x x = Y (9) From (9), by the least squares method, the arameters vector P x is being determined: T T x x x x (8) P = ( U U ) U Y () The structure of the fuy system from Figure is similar to the structure of the RBF network. In [3] is given the detailed comarison of these two systems. It was concluded that there exist the functional equivalency if: The number of invisible units of the RBF network is equal to number of linguistic rules in fuy system For the fuy system of Figure holds f i = const The Gaussian functions are selected as the membershi rimary fuy sets for the system of Figure For determination of the T-norm in fuy systems, one uses the algebraic roduct The activation threshold of the outut neuron for the RBF network is ero. Since the structure of the fuy system of Figure is similar to the structure of the RBF neural network, and taking into account that there exist functional equivalency, the adative fuy system is frequently called the neuro-fuy system.

7 Alication of the Takagi-Sugeno Fuy Controller for Solving the Robots' Inverse Kinematics Problem APPLICATION OF THE TAKAGI-SUGENO FUZZY CONTROLLER TO SOLVING THE PROBLEM OF ROBOTS' INVERSE KINEMATICS In aer [5] the FLC is alied to solving the roblem of inverse kinematics. In all controllers, the fuy reasoning of the first tye was alied, where the Mamdani minimum rule is alied as a function in the imlication hase. The biggest deficiency of such FLC structures is long comutational time. Esecially roblematic is definition of the base rule and selection of the membershi functions of the rimary fuy sets. In this aer to solving of the IK is alied the Takagi-Sugeno controller shown in Figure. Simulations were done for the robot shown in Figure (the R R y R y configuration). The robot segments' lengths are: l =.3 m, l = l 3 = m. It is taken that there are constraints in joints (θ = θ = θ 3 = [ π/, π/]). The inut and the outut variables of the FLC for solving the IK are shown in Figure 3. Simulations were erformed with the rogram ackage MATLAB, by using the fuy toolbox. θ 3 θ l L 3 l y x θ x Fig.. The RRR robot of minimal configuration FLC θ x FLC θ x FLC 3 θ 3 Fig. 3. Illustration of the inut and the outut variables of the Takagi-Sugeno FLCs for solving the IK of the R R y R y robot. In the first examle, the grier was moving along the straight-line trajectory, from the oint A(.4.7) to oint B( ), with the triangular velocity rofile. The time taken for erforming the motion is t max = s. In this examle by the data set for training of the FLCs is enhanced only one straight-line trajectory. The learning seeds η a, η b and η c were taken according to [4] in the form: k η a =η b =η c = ε ε ε + + aq bq c q

8 46 V. RANKOVIĆ, I. NIKOLIĆ where k is the ste. The ste sie is variable, and so does the convergence rate. According to [], if k is small, the convergence is slow, and several iteraqtions would be necessary, to achieve satisfactory accuracy. If k is big, the convergence would, at the beginning of the learning rocess, be fast, but the algorithm will oscillate. The ste k changes according to the following rules: If the error is decreasing in the four consecutive eochs, k increases for %. If the error in one eoch decreases, then in the next one k decreases for %. In the first examle for the FLC is k =., and for FLC and FLC 3 is k =.. The fuy artitioning of the inut variables of the FLC is realied by selection of rimary fuy sets. Selected are the Gaussian membershi functions, since the best results are achieved with them. The total number of the fuy rules is. In the learning rocess the membershi functions arameters of the rimary fuy sets () were adated as well as arameters of the Takagi-Sugeno controller (). In Figure 4 are resented the membershi functions of the rimary fuy sets of variables x and y before learning, and in Figure 5 are resented the membershi functions after learning. µ(x) A A µ(y) A A µ(x) x y Fig. 4. Membershi functions of variables x and y before learning. A A µ(y) A A x y Fig. 5. Membershi functions of variables x and y after learning the FLC. In Figure 6 is resented the variation of error, given by exression (8), as a function of the number of eochs. E 8 X Eoch Fig. 6. Error variation during the learning rocess of the FLC. Parameters of the fi functions, i =,4, are given in Table.

9 Alication of the Takagi-Sugeno Fuy Controller for Solving the Robots' Inverse Kinematics Problem 47 In FLC to each inut variable were assigned two membershi functions. The controller has total of 4 rules. The total number of arameters, which are being adated is 4, out of which are the remise arameters and are the consequences arameters. In Figure 7 are shown the membershi functions of the rimary fuy sets of variables x i and before learning, and in Figure 8 are shown the membershi functions after learning. The variation of error with the number of eochs is shown in Figure 9. µ(x ) µ() A A A A µ(x ) x Fig. 7. Membershi functions of variables x i and before learning. A A A µ() A x Fig. 8. Membershi functions of variables x i and after learning the FLC. E x Eoch Fig. 9. Error variation during the learning rocess of the FLC. Parameters of the fi functions, i =,4, for FLC are given in Table. FLC 3 has 4 rules (two membershi functions were assigned to each inut). The membershi functions of the rimary fuy sets of variables x i and, rior to the adatation rocess, are shown in Figure 7, while the figure shows them after the learning rocess. In Figure is illustrated the variation of error, given with exression (8) as a function of the number of eochs. µ(x ) A A µ() A A x Fig.. Membershi functions of variables x i and after learning the FLC 3.

10 48 V. RANKOVIĆ, I. NIKOLIĆ E 3 x Eoch Fig.. Error variation during the learning rocess of the FLC 3. Parameters of the fi functions, i =,4, for FLC 3 are given in Table. In Figure is given the variation of the interior coordinates during executing the working task, and in Figure 3 is given the variation of the total error of tracking the straight-line trajectory. Table. Parameters of consequences of the Takagi-Sugeno controllers FLC FLC FLC c c c c θ [ ] θ i rad.4. θ θ. θ T s [ ] Fig.. Variation of the internal coordinates along the trajectory From Figure 3 can be seen that the maximum total error of the straight-line trajectory tracking, in the first examle is less than mm. The simulation results show

11 Alication of the Takagi-Sugeno Fuy Controller for Solving the Robots' Inverse Kinematics Problem 49 that the alication of the neuro-fuy controllers, with the membershi functions, shown in Figures 5, 8 and, and with arameters of the consequences functions given in Table, give the satisfactory results. The FLCs from this examle have smaller number of linguistic rules than the number of the invisible units of the RBF networks, generated for solving the same roblem, shown in [6]. -5 x 5.5 em [ ] T s Fig. 3. The total tracking error [] In the second examle the grier was moving along the straight-line trajectory, form the oint A( ) to oint B(.6.3.3), with the seed equal to v =. m/s, and the training data set enhanced the art of the working area x = [;.7], y = [; ] and = [;,8]. The training set redicted that the trajectories can be realied both over the ositive or negative values of the internal coordinate θ 3. The FLC determines the internal coordinate θ, the FLC θ for ositive values of θ 3, and FLC 3 generates the coordinate θ for negative values of θ 3, and FLC 4 θ 3. Selection of the FLC, which would determine the coordinate θ, is realied based on the robot's initial osition. In FLC to each inut variable are assigned three membershi functions. The controller has total of 9 rules. The total number of arameters, which are being adated is 45, out of which 8 are the remise arameters, and 7 are the consequences arameters. Fuy artitioning of the inut variables FLC, FLC 3 and FLC 4 is realied by selection of 5 rimary fuy sets for each variable. The membershi functions are selected as Gaussian. The total number of the fuy rules of each controller is 5. In the learning rocess, the arameters of the membershi functions of the rimary fuy sets were adated (3) as well as the arameters of the consequences of the Takagi-Sugeno controller (75). In Figures 4 and 5 are shown the membershi functions of the rimary fuy sets of variables x and y, for FLC, and variables x and for FLC, FLC 3 and FLC 4, rior to learning. In Figures 6 to 9 are resented the membershi functions after learning. µ(x) A A A 3 µ(y)a A A x y Fig. 4. Membershi functions of variables x and y before learning.

12 5 V. RANKOVIĆ, I. NIKOLIĆ µ(x ) A A A 3 A 4 A 5 µ() A A A 3 A 4 A x Fig. 5. Membershi functions of variables x and i before learning. µ(x) A A A 3 µ(y)a A A x y Figure 6. Membershi functions of variables x and y after learning the FLC. In Figures to 3 are shown error variations as a function of number of eochs for all four networks. µ(x ) A A A 3 A 4 A 5 µ() A A A 3 A 4 A x Fig. 7. Membershi functions of variables x and i after learning the FLC. µ(x ) A A A 3 A 4 A 5 µ() A A A 3 A 4 A x Fig. 8. Membershi functions of variables x and i after learning the FLC 3. µ(x ) A A A 3 A 4 A 5 µ() A A A 3 A 4 A x Fig. 9. Membershi functions of variables x and i after learning the FLC 4.

13 Alication of the Takagi-Sugeno Fuy Controller for Solving the Robots' Inverse Kinematics Problem 5 E X Eoch Fig.. Error variation during the learning rocess of the FLC. E X Eoch Fig.. Error variation during the learning rocess of the FLC. E 3 3 X Eoch Fig.. Error variation during the learning rocess of the FLC 3. E 4 6 X Eoch Fig. 3. Error variation during the learning rocess of the FLC 4. Parameters of functions fi, for FLC, FLC, FLC 3 and FLC 4 are given in Table. In Figure 4 is resented variation of the internal coordinates during the task execution, and in Figure 5 is given the total error variation of the straight-line trajectory tracking.

14 5 V. RANKOVIĆ, I. NIKOLIĆ θ i.8 [ rad ].6 θ θ 3.4. θ T s [ ] Fig. 4. Variation of the internal coordinates along the trajectory Table. Consequences arameters FLC FLC FLC 3 FLC c c c c " " " " " c c c Maximum total error of the trajectory tracking in the second examle is less than. mm. -3 x. em [ ] T s [] Fig. 5. The total tracking error

15 Alication of the Takagi-Sugeno Fuy Controller for Solving the Robots' Inverse Kinematics Problem CONCLUSION Results of both simulations, resented in this aer, show that the alication of the neuro-fuy system to solving IK gives satisfactory results. Advantage of alication of the Takagi-Sugeno controllers, imlemented within the adative network, related to solving IK by alication od the FLCs roosed in [4], lies in the fact that the ste between inter-oints on the straight-line trajectory, does not affect the accuracy of tracking. The tracking error is smaller for the case of solving the IK by RBF neural networks []. However, the main deficiency of networks trained by data sets for learning, which enhance the art of the working sace, is the fact that their hidden layers contain large number of neurons. The FLCs from the second examle have smaller number of linguistic rules than the number of the invisible units of the RBF networks generated for solving the same roblem. For solving the IK during the straight-line motion along the ath from oint A( ) to oint B(.6.3.3) with the velocity v =. m/s, the number of neurons of the hidden layer of the first RBF network is, while the number of rules FLC is 9. The RBF and RBF 3 have 86 and 74 neurons, resectively, and FLC and FLC 3 have 5 rules, each. Simulated FLCs have larger number of arameters, which are being adated. Weights of connections between invisible units and oututs, in RBF networks are constant values, while the consequences functions for Takagi-Sugeno controllers are linear functions of inuts into the controller. In realiation of fast trajectories, advantage for solving IK should be given to Takagi- Sugeno controllers with resect to the RBF neural networks [6]. Due to large number of neurons, RBF neural networks cannot realie maing from the external to internal coordinates in real time for high grier velocities. If it is necessary to realie trajectory with an error less than mm, then the data set for training should enhance only trajectories, which the grier tracks during the task execution. REFERENCES. Godjevac, J., (997) "A method of the design of neuro-fuy controllers; an alication in robot learning," Doctorial dissertation, EPFL, Lausanne, Switerland.. Jang, S. R., (993) "ANFIS: Adative-Network-Based Fuy Inference System," IEEE Trans. on Systems, Man and Cybernetics, vol. 3, no. 3, Jang, S. R., C. T. Sun, (993) "Functional Equivalence between Radial basis Function Networks and Fuy Inference Systems," IEEE Trans. on Neural Networks, vol. 4, no., Nedungadi, A., (995) "Alication of Fuy Logic to Solve the Robot Inverse Kinematic Problem," in Proc. of Fourth World Conference Robotics Research, Pittsburgh, PA, USA, Nikolić, I., M. Janković, (998) "Fuy Controller for Solving the Inverse Kinematics of Industrial Robots", Proceedings of the XXVII ETRAN Conference, Vrnjačka Banja, Volume 3, Ranković, V., I., Nikolić, () "Solving of the Robots' Inverse Kinematics by Two-Layer Neural Network", Proceedings of XLV ETRAN Conference, Bukovička Banja, Volume 3,. 7-3.

16 54 V. RANKOVIĆ, I. NIKOLIĆ PRIMENA TAKAGI-SUGENO FUZZY KONTROLERA ZA RE[AVANJE PROBLEMA INVERZNE KINEMATIKE ROBOTA Vesna Rankovi}, Ilija Nikoli} Za reslikavanje i soljašnjih u unutrašnje koordinate je korišćen Takagi-Sugeno kontroler, imlementiran u adativnu mre`u, koji je o arhitekturi sličan RBF neuronskoj mre`i, a se takodje naiva neuro fuy sistemom. Tokom rocesa u~enja su adatirani arametri funkcija riadnosti rimarnih fuy skuova i arametri osledica Takagi-Sugeno kontrolera. Prikaani reultati se odnose na ra}enje ravolinijske trajektorije sa konstantnom brinom hvataljke i trougaonim rofilom brine.

Feedback-error control

Feedback-error control Chater 4 Feedback-error control 4.1 Introduction This chater exlains the feedback-error (FBE) control scheme originally described by Kawato [, 87, 8]. FBE is a widely used neural network based controller

More information

Uncertainty Modeling with Interval Type-2 Fuzzy Logic Systems in Mobile Robotics

Uncertainty Modeling with Interval Type-2 Fuzzy Logic Systems in Mobile Robotics Uncertainty Modeling with Interval Tye-2 Fuzzy Logic Systems in Mobile Robotics Ondrej Linda, Student Member, IEEE, Milos Manic, Senior Member, IEEE bstract Interval Tye-2 Fuzzy Logic Systems (IT2 FLSs)

More information

Solved Problems. (a) (b) (c) Figure P4.1 Simple Classification Problems First we draw a line between each set of dark and light data points.

Solved Problems. (a) (b) (c) Figure P4.1 Simple Classification Problems First we draw a line between each set of dark and light data points. Solved Problems Solved Problems P Solve the three simle classification roblems shown in Figure P by drawing a decision boundary Find weight and bias values that result in single-neuron ercetrons with the

More information

Radial Basis Function Networks: Algorithms

Radial Basis Function Networks: Algorithms Radial Basis Function Networks: Algorithms Introduction to Neural Networks : Lecture 13 John A. Bullinaria, 2004 1. The RBF Maing 2. The RBF Network Architecture 3. Comutational Power of RBF Networks 4.

More information

Fuzzy Automata Induction using Construction Method

Fuzzy Automata Induction using Construction Method Journal of Mathematics and Statistics 2 (2): 395-4, 26 ISSN 1549-3644 26 Science Publications Fuzzy Automata Induction using Construction Method 1,2 Mo Zhi Wen and 2 Wan Min 1 Center of Intelligent Control

More information

FUZZY CONTROL. Khurshid Ahmad, Professor of Computer Science, Department of Computer Science Trinity College, Dublin-2, IRELAND Nov 15-16th, 2011

FUZZY CONTROL. Khurshid Ahmad, Professor of Computer Science, Department of Computer Science Trinity College, Dublin-2, IRELAND Nov 15-16th, 2011 : Mamdani & akagi-sugeno Controllers Khurshid Ahmad, Professor of Comuter Science, Deartment of Comuter Science rinit College, Dublin-, IRELAND Nov 5-6th, 0 htts://www.cs.tcd.ie/khurshid.ahmad/eaching/eaching.html

More information

Design of NARMA L-2 Control of Nonlinear Inverted Pendulum

Design of NARMA L-2 Control of Nonlinear Inverted Pendulum International Research Journal of Alied and Basic Sciences 016 Available online at www.irjabs.com ISSN 51-838X / Vol, 10 (6): 679-684 Science Exlorer Publications Design of NARMA L- Control of Nonlinear

More information

Multivariable Generalized Predictive Scheme for Gas Turbine Control in Combined Cycle Power Plant

Multivariable Generalized Predictive Scheme for Gas Turbine Control in Combined Cycle Power Plant Multivariable Generalized Predictive Scheme for Gas urbine Control in Combined Cycle Power Plant L.X.Niu and X.J.Liu Deartment of Automation North China Electric Power University Beiing, China, 006 e-mail

More information

Distributed Rule-Based Inference in the Presence of Redundant Information

Distributed Rule-Based Inference in the Presence of Redundant Information istribution Statement : roved for ublic release; distribution is unlimited. istributed Rule-ased Inference in the Presence of Redundant Information June 8, 004 William J. Farrell III Lockheed Martin dvanced

More information

Combining Logistic Regression with Kriging for Mapping the Risk of Occurrence of Unexploded Ordnance (UXO)

Combining Logistic Regression with Kriging for Mapping the Risk of Occurrence of Unexploded Ordnance (UXO) Combining Logistic Regression with Kriging for Maing the Risk of Occurrence of Unexloded Ordnance (UXO) H. Saito (), P. Goovaerts (), S. A. McKenna (2) Environmental and Water Resources Engineering, Deartment

More information

Use of Transformations and the Repeated Statement in PROC GLM in SAS Ed Stanek

Use of Transformations and the Repeated Statement in PROC GLM in SAS Ed Stanek Use of Transformations and the Reeated Statement in PROC GLM in SAS Ed Stanek Introduction We describe how the Reeated Statement in PROC GLM in SAS transforms the data to rovide tests of hyotheses of interest.

More information

MODELING THE RELIABILITY OF C4ISR SYSTEMS HARDWARE/SOFTWARE COMPONENTS USING AN IMPROVED MARKOV MODEL

MODELING THE RELIABILITY OF C4ISR SYSTEMS HARDWARE/SOFTWARE COMPONENTS USING AN IMPROVED MARKOV MODEL Technical Sciences and Alied Mathematics MODELING THE RELIABILITY OF CISR SYSTEMS HARDWARE/SOFTWARE COMPONENTS USING AN IMPROVED MARKOV MODEL Cezar VASILESCU Regional Deartment of Defense Resources Management

More information

LIMITATIONS OF RECEPTRON. XOR Problem The failure of the perceptron to successfully simple problem such as XOR (Minsky and Papert).

LIMITATIONS OF RECEPTRON. XOR Problem The failure of the perceptron to successfully simple problem such as XOR (Minsky and Papert). LIMITATIONS OF RECEPTRON XOR Problem The failure of the ercetron to successfully simle roblem such as XOR (Minsky and Paert). x y z x y z 0 0 0 0 0 0 Fig. 4. The exclusive-or logic symbol and function

More information

Robust Performance Design of PID Controllers with Inverse Multiplicative Uncertainty

Robust Performance Design of PID Controllers with Inverse Multiplicative Uncertainty American Control Conference on O'Farrell Street San Francisco CA USA June 9 - July Robust Performance Design of PID Controllers with Inverse Multilicative Uncertainty Tooran Emami John M Watkins Senior

More information

and INVESTMENT DECISION-MAKING

and INVESTMENT DECISION-MAKING CONSISTENT INTERPOLATIVE FUZZY LOGIC and INVESTMENT DECISION-MAKING Darko Kovačević, č Petar Sekulić and dal Aleksandar Rakićević National bank of Serbia Belgrade, Jun 23th, 20 The structure of the resentation

More information

State Estimation with ARMarkov Models

State Estimation with ARMarkov Models Deartment of Mechanical and Aerosace Engineering Technical Reort No. 3046, October 1998. Princeton University, Princeton, NJ. State Estimation with ARMarkov Models Ryoung K. Lim 1 Columbia University,

More information

General Linear Model Introduction, Classes of Linear models and Estimation

General Linear Model Introduction, Classes of Linear models and Estimation Stat 740 General Linear Model Introduction, Classes of Linear models and Estimation An aim of scientific enquiry: To describe or to discover relationshis among events (variables) in the controlled (laboratory)

More information

Analysis of the Effect of Number of Knots in a Trajectory on Motion Characteristics of a 3R Planar Manipulator

Analysis of the Effect of Number of Knots in a Trajectory on Motion Characteristics of a 3R Planar Manipulator Analysis of the Effect of Number of Knots in a Trajectory on Motion Characteristics of a Planar Maniulator Suarno Bhattacharyya & Tarun Kanti Naskar Mechanical Engineering Deartment, Jadavur University,

More information

Feedback-Based Iterative Learning Control for MIMO LTI Systems

Feedback-Based Iterative Learning Control for MIMO LTI Systems International Journal of Control, Feedback-Based Automation, Iterative and Systems, Learning vol. Control 6, no., for. MIMO 69-77, LTI Systems Aril 8 69 Feedback-Based Iterative Learning Control for MIMO

More information

Shadow Computing: An Energy-Aware Fault Tolerant Computing Model

Shadow Computing: An Energy-Aware Fault Tolerant Computing Model Shadow Comuting: An Energy-Aware Fault Tolerant Comuting Model Bryan Mills, Taieb Znati, Rami Melhem Deartment of Comuter Science University of Pittsburgh (bmills, znati, melhem)@cs.itt.edu Index Terms

More information

A Method of Setting the Penalization Constants in the Suboptimal Linear Quadratic Tracking Method

A Method of Setting the Penalization Constants in the Suboptimal Linear Quadratic Tracking Method XXVI. ASR '21 Seminar, Instruments and Control, Ostrava, Aril 26-27, 21 Paer 57 A Method of Setting the Penalization Constants in the Subotimal Linear Quadratic Tracking Method PERŮTKA, Karel Ing., Deartment

More information

COMPARISON OF VARIOUS OPTIMIZATION TECHNIQUES FOR DESIGN FIR DIGITAL FILTERS

COMPARISON OF VARIOUS OPTIMIZATION TECHNIQUES FOR DESIGN FIR DIGITAL FILTERS NCCI 1 -National Conference on Comutational Instrumentation CSIO Chandigarh, INDIA, 19- March 1 COMPARISON OF VARIOUS OPIMIZAION ECHNIQUES FOR DESIGN FIR DIGIAL FILERS Amanjeet Panghal 1, Nitin Mittal,Devender

More information

Dynamic System Eigenvalue Extraction using a Linear Echo State Network for Small-Signal Stability Analysis a Novel Application

Dynamic System Eigenvalue Extraction using a Linear Echo State Network for Small-Signal Stability Analysis a Novel Application Dynamic System Eigenvalue Extraction using a Linear Echo State Network for Small-Signal Stability Analysis a Novel Alication Jiaqi Liang, Jing Dai, Ganesh K. Venayagamoorthy, and Ronald G. Harley Abstract

More information

Computer arithmetic. Intensive Computation. Annalisa Massini 2017/2018

Computer arithmetic. Intensive Computation. Annalisa Massini 2017/2018 Comuter arithmetic Intensive Comutation Annalisa Massini 7/8 Intensive Comutation - 7/8 References Comuter Architecture - A Quantitative Aroach Hennessy Patterson Aendix J Intensive Comutation - 7/8 3

More information

On Fractional Predictive PID Controller Design Method Emmanuel Edet*. Reza Katebi.**

On Fractional Predictive PID Controller Design Method Emmanuel Edet*. Reza Katebi.** On Fractional Predictive PID Controller Design Method Emmanuel Edet*. Reza Katebi.** * echnology and Innovation Centre, Level 4, Deartment of Electronic and Electrical Engineering, University of Strathclyde,

More information

The Motion Path Study of Measuring Robot Based on Variable Universe Fuzzy Control

The Motion Path Study of Measuring Robot Based on Variable Universe Fuzzy Control MATEC Web of Conferences 95, 8 (27) DOI:.5/ matecconf/27958 ICMME 26 The Motion Path Study of Measuring Robot Based on Variable Universe Fuzzy Control Ma Guoqing, Yu Zhenglin, Cao Guohua, Zhang Ruoyan

More information

Recent Developments in Multilayer Perceptron Neural Networks

Recent Developments in Multilayer Perceptron Neural Networks Recent Develoments in Multilayer Percetron eural etworks Walter H. Delashmit Lockheed Martin Missiles and Fire Control Dallas, Texas 75265 walter.delashmit@lmco.com walter.delashmit@verizon.net Michael

More information

Bayesian Model Averaging Kriging Jize Zhang and Alexandros Taflanidis

Bayesian Model Averaging Kriging Jize Zhang and Alexandros Taflanidis HIPAD LAB: HIGH PERFORMANCE SYSTEMS LABORATORY DEPARTMENT OF CIVIL AND ENVIRONMENTAL ENGINEERING AND EARTH SCIENCES Bayesian Model Averaging Kriging Jize Zhang and Alexandros Taflanidis Why use metamodeling

More information

STABILITY ANALYSIS TOOL FOR TUNING UNCONSTRAINED DECENTRALIZED MODEL PREDICTIVE CONTROLLERS

STABILITY ANALYSIS TOOL FOR TUNING UNCONSTRAINED DECENTRALIZED MODEL PREDICTIVE CONTROLLERS STABILITY ANALYSIS TOOL FOR TUNING UNCONSTRAINED DECENTRALIZED MODEL PREDICTIVE CONTROLLERS Massimo Vaccarini Sauro Longhi M. Reza Katebi D.I.I.G.A., Università Politecnica delle Marche, Ancona, Italy

More information

Oil Temperature Control System PID Controller Algorithm Analysis Research on Sliding Gear Reducer

Oil Temperature Control System PID Controller Algorithm Analysis Research on Sliding Gear Reducer Key Engineering Materials Online: 2014-08-11 SSN: 1662-9795, Vol. 621, 357-364 doi:10.4028/www.scientific.net/kem.621.357 2014 rans ech Publications, Switzerland Oil emerature Control System PD Controller

More information

E( x ) [b(n) - a(n, m)x(m) ]

E( x ) [b(n) - a(n, m)x(m) ] Homework #, EE5353. An XOR network has two inuts, one hidden unit, and one outut. It is fully connected. Gie the network's weights if the outut unit has a ste actiation and the hidden unit actiation is

More information

2-D Analysis for Iterative Learning Controller for Discrete-Time Systems With Variable Initial Conditions Yong FANG 1, and Tommy W. S.

2-D Analysis for Iterative Learning Controller for Discrete-Time Systems With Variable Initial Conditions Yong FANG 1, and Tommy W. S. -D Analysis for Iterative Learning Controller for Discrete-ime Systems With Variable Initial Conditions Yong FANG, and ommy W. S. Chow Abstract In this aer, an iterative learning controller alying to linear

More information

Minimax Design of Nonnegative Finite Impulse Response Filters

Minimax Design of Nonnegative Finite Impulse Response Filters Minimax Design of Nonnegative Finite Imulse Resonse Filters Xiaoing Lai, Anke Xue Institute of Information and Control Hangzhou Dianzi University Hangzhou, 3118 China e-mail: laix@hdu.edu.cn; akxue@hdu.edu.cn

More information

Nonlinear Static Analysis of Cable Net Structures by Using Newton-Raphson Method

Nonlinear Static Analysis of Cable Net Structures by Using Newton-Raphson Method Nonlinear Static Analysis of Cable Net Structures by Using Newton-Rahson Method Sayed Mahdi Hazheer Deartment of Civil Engineering University Selangor (UNISEL) Selangor, Malaysia hazheer.ma@gmail.com Abstract

More information

Multilayer Perceptron Neural Network (MLPs) For Analyzing the Properties of Jordan Oil Shale

Multilayer Perceptron Neural Network (MLPs) For Analyzing the Properties of Jordan Oil Shale World Alied Sciences Journal 5 (5): 546-552, 2008 ISSN 1818-4952 IDOSI Publications, 2008 Multilayer Percetron Neural Network (MLPs) For Analyzing the Proerties of Jordan Oil Shale 1 Jamal M. Nazzal, 2

More information

On Line Parameter Estimation of Electric Systems using the Bacterial Foraging Algorithm

On Line Parameter Estimation of Electric Systems using the Bacterial Foraging Algorithm On Line Parameter Estimation of Electric Systems using the Bacterial Foraging Algorithm Gabriel Noriega, José Restreo, Víctor Guzmán, Maribel Giménez and José Aller Universidad Simón Bolívar Valle de Sartenejas,

More information

GENERATING FUZZY RULES FOR PROTEIN CLASSIFICATION E. G. MANSOORI, M. J. ZOLGHADRI, S. D. KATEBI, H. MOHABATKAR, R. BOOSTANI AND M. H.

GENERATING FUZZY RULES FOR PROTEIN CLASSIFICATION E. G. MANSOORI, M. J. ZOLGHADRI, S. D. KATEBI, H. MOHABATKAR, R. BOOSTANI AND M. H. Iranian Journal of Fuzzy Systems Vol. 5, No. 2, (2008). 21-33 GENERATING FUZZY RULES FOR PROTEIN CLASSIFICATION E. G. MANSOORI, M. J. ZOLGHADRI, S. D. KATEBI, H. MOHABATKAR, R. BOOSTANI AND M. H. SADREDDINI

More information

The Noise Power Ratio - Theory and ADC Testing

The Noise Power Ratio - Theory and ADC Testing The Noise Power Ratio - Theory and ADC Testing FH Irons, KJ Riley, and DM Hummels Abstract This aer develos theory behind the noise ower ratio (NPR) testing of ADCs. A mid-riser formulation is used for

More information

Numerical Methods for Particle Tracing in Vector Fields

Numerical Methods for Particle Tracing in Vector Fields On-Line Visualization Notes Numerical Methods for Particle Tracing in Vector Fields Kenneth I. Joy Visualization and Grahics Research Laboratory Deartment of Comuter Science University of California, Davis

More information

Metrics Performance Evaluation: Application to Face Recognition

Metrics Performance Evaluation: Application to Face Recognition Metrics Performance Evaluation: Alication to Face Recognition Naser Zaeri, Abeer AlSadeq, and Abdallah Cherri Electrical Engineering Det., Kuwait University, P.O. Box 5969, Safat 6, Kuwait {zaery, abeer,

More information

A Comparison between Biased and Unbiased Estimators in Ordinary Least Squares Regression

A Comparison between Biased and Unbiased Estimators in Ordinary Least Squares Regression Journal of Modern Alied Statistical Methods Volume Issue Article 7 --03 A Comarison between Biased and Unbiased Estimators in Ordinary Least Squares Regression Ghadban Khalaf King Khalid University, Saudi

More information

AI*IA 2003 Fusion of Multiple Pattern Classifiers PART III

AI*IA 2003 Fusion of Multiple Pattern Classifiers PART III AI*IA 23 Fusion of Multile Pattern Classifiers PART III AI*IA 23 Tutorial on Fusion of Multile Pattern Classifiers by F. Roli 49 Methods for fusing multile classifiers Methods for fusing multile classifiers

More information

Towards understanding the Lorenz curve using the Uniform distribution. Chris J. Stephens. Newcastle City Council, Newcastle upon Tyne, UK

Towards understanding the Lorenz curve using the Uniform distribution. Chris J. Stephens. Newcastle City Council, Newcastle upon Tyne, UK Towards understanding the Lorenz curve using the Uniform distribution Chris J. Stehens Newcastle City Council, Newcastle uon Tyne, UK (For the Gini-Lorenz Conference, University of Siena, Italy, May 2005)

More information

Power System Reactive Power Optimization Based on Fuzzy Formulation and Interior Point Filter Algorithm

Power System Reactive Power Optimization Based on Fuzzy Formulation and Interior Point Filter Algorithm Energy and Power Engineering, 203, 5, 693-697 doi:0.4236/ee.203.54b34 Published Online July 203 (htt://www.scir.org/journal/ee Power System Reactive Power Otimization Based on Fuzzy Formulation and Interior

More information

ANALYTIC APPROXIMATE SOLUTIONS FOR FLUID-FLOW IN THE PRESENCE OF HEAT AND MASS TRANSFER

ANALYTIC APPROXIMATE SOLUTIONS FOR FLUID-FLOW IN THE PRESENCE OF HEAT AND MASS TRANSFER Kilicman, A., et al.: Analytic Aroximate Solutions for Fluid-Flow in the Presence... THERMAL SCIENCE: Year 8, Vol., Sul.,. S59-S64 S59 ANALYTIC APPROXIMATE SOLUTIONS FOR FLUID-FLOW IN THE PRESENCE OF HEAT

More information

ALTERNATIVE SOLUTION TO THE QUARTIC EQUATION by Farid A. Chouery 1, P.E. 2006, All rights reserved

ALTERNATIVE SOLUTION TO THE QUARTIC EQUATION by Farid A. Chouery 1, P.E. 2006, All rights reserved ALTERNATIVE SOLUTION TO THE QUARTIC EQUATION b Farid A. Chouer, P.E. 006, All rights reserved Abstract A new method to obtain a closed form solution of the fourth order olnomial equation is roosed in this

More information

Passive Identification is Non Stationary Objects With Closed Loop Control

Passive Identification is Non Stationary Objects With Closed Loop Control IOP Conerence Series: Materials Science and Engineering PAPER OPEN ACCESS Passive Identiication is Non Stationary Obects With Closed Loo Control To cite this article: Valeriy F Dyadik et al 2016 IOP Con.

More information

MODELLING, SIMULATION AND ROBUST ANALYSIS OF THE TEMPERATURE PROCESS CONTROL

MODELLING, SIMULATION AND ROBUST ANALYSIS OF THE TEMPERATURE PROCESS CONTROL The 6 th edition of the Interdiscilinarity in Engineering International Conference Petru Maior University of Tîrgu Mureş, Romania, 2012 MODELLING, SIMULATION AND ROBUST ANALYSIS OF THE TEMPERATURE PROCESS

More information

Position Control of Induction Motors by Exact Feedback Linearization *

Position Control of Induction Motors by Exact Feedback Linearization * BULGARIAN ACADEMY OF SCIENCES CYBERNETICS AND INFORMATION TECHNOLOGIES Volume 8 No Sofia 008 Position Control of Induction Motors by Exact Feedback Linearization * Kostadin Kostov Stanislav Enev Farhat

More information

Research Article Research on Evaluation Indicator System and Methods of Food Network Marketing Performance

Research Article Research on Evaluation Indicator System and Methods of Food Network Marketing Performance Advance Journal of Food Science and Technology 7(0: 80-84 205 DOI:0.9026/afst.7.988 ISSN: 2042-4868; e-issn: 2042-4876 205 Maxwell Scientific Publication Cor. Submitted: October 7 204 Acceted: December

More information

Research of PMU Optimal Placement in Power Systems

Research of PMU Optimal Placement in Power Systems Proceedings of the 5th WSEAS/IASME Int. Conf. on SYSTEMS THEORY and SCIENTIFIC COMPUTATION, Malta, Setember 15-17, 2005 (38-43) Research of PMU Otimal Placement in Power Systems TIAN-TIAN CAI, QIAN AI

More information

FE FORMULATIONS FOR PLASTICITY

FE FORMULATIONS FOR PLASTICITY G These slides are designed based on the book: Finite Elements in Plasticity Theory and Practice, D.R.J. Owen and E. Hinton, 1970, Pineridge Press Ltd., Swansea, UK. 1 Course Content: A INTRODUCTION AND

More information

Session 5: Review of Classical Astrodynamics

Session 5: Review of Classical Astrodynamics Session 5: Review of Classical Astrodynamics In revious lectures we described in detail the rocess to find the otimal secific imulse for a articular situation. Among the mission requirements that serve

More information

A Simple Weight Decay Can Improve. Abstract. It has been observed in numerical simulations that a weight decay can improve

A Simple Weight Decay Can Improve. Abstract. It has been observed in numerical simulations that a weight decay can improve In Advances in Neural Information Processing Systems 4, J.E. Moody, S.J. Hanson and R.P. Limann, eds. Morgan Kaumann Publishers, San Mateo CA, 1995,. 950{957. A Simle Weight Decay Can Imrove Generalization

More information

ADAPTIVE NEURO-FUZZY INFERENCE SYSTEMS

ADAPTIVE NEURO-FUZZY INFERENCE SYSTEMS ADAPTIVE NEURO-FUZZY INFERENCE SYSTEMS RBFN and TS systems Equivalent if the following hold: Both RBFN and TS use same aggregation method for output (weighted sum or weighted average) Number of basis functions

More information

Determining Momentum and Energy Corrections for g1c Using Kinematic Fitting

Determining Momentum and Energy Corrections for g1c Using Kinematic Fitting CLAS-NOTE 4-17 Determining Momentum and Energy Corrections for g1c Using Kinematic Fitting Mike Williams, Doug Alegate and Curtis A. Meyer Carnegie Mellon University June 7, 24 Abstract We have used the

More information

MULTI-CHANNEL PARAMETRIC ESTIMATOR FAST BLOCK MATRIX INVERSES

MULTI-CHANNEL PARAMETRIC ESTIMATOR FAST BLOCK MATRIX INVERSES MULTI-CANNEL ARAMETRIC ESTIMATOR FAST BLOCK MATRIX INVERSES S Lawrence Marle Jr School of Electrical Engineering and Comuter Science Oregon State University Corvallis, OR 97331 Marle@eecsoregonstateedu

More information

MODEL-BASED MULTIPLE FAULT DETECTION AND ISOLATION FOR NONLINEAR SYSTEMS

MODEL-BASED MULTIPLE FAULT DETECTION AND ISOLATION FOR NONLINEAR SYSTEMS MODEL-BASED MULIPLE FAUL DEECION AND ISOLAION FOR NONLINEAR SYSEMS Ivan Castillo, and homas F. Edgar he University of exas at Austin Austin, X 78712 David Hill Chemstations Houston, X 77009 Abstract A

More information

Covariance Matrix Estimation for Reinforcement Learning

Covariance Matrix Estimation for Reinforcement Learning Covariance Matrix Estimation for Reinforcement Learning Tomer Lancewicki Deartment of Electrical Engineering and Comuter Science University of Tennessee Knoxville, TN 37996 tlancewi@utk.edu Itamar Arel

More information

Lower Confidence Bound for Process-Yield Index S pk with Autocorrelated Process Data

Lower Confidence Bound for Process-Yield Index S pk with Autocorrelated Process Data Quality Technology & Quantitative Management Vol. 1, No.,. 51-65, 15 QTQM IAQM 15 Lower onfidence Bound for Process-Yield Index with Autocorrelated Process Data Fu-Kwun Wang * and Yeneneh Tamirat Deartment

More information

Comparative study between the conventional regulators and fuzzy logic controller: application on the induction machine

Comparative study between the conventional regulators and fuzzy logic controller: application on the induction machine International Journal of Sciences and Techniques of Automatic control & comuter engineering IJ-STA, Volume 1, N, December 007,. 196 1. Comarative study between the conventional regulators and fuzzy logic

More information

José Alberto Mauricio. Instituto Complutense de Análisis Económico. Universidad Complutense de Madrid. Campus de Somosaguas Madrid - SPAIN

José Alberto Mauricio. Instituto Complutense de Análisis Económico. Universidad Complutense de Madrid. Campus de Somosaguas Madrid - SPAIN A CORRECTED ALGORITHM FOR COMPUTING THE THEORETICAL AUTOCOVARIANCE MATRICES OF A VECTOR ARMA MODEL José Alberto Mauricio Instituto Comlutense de Análisis Económico Universidad Comlutense de Madrid Camus

More information

Detection Algorithm of Particle Contamination in Reticle Images with Continuous Wavelet Transform

Detection Algorithm of Particle Contamination in Reticle Images with Continuous Wavelet Transform Detection Algorithm of Particle Contamination in Reticle Images with Continuous Wavelet Transform Chaoquan Chen and Guoing Qiu School of Comuter Science and IT Jubilee Camus, University of Nottingham Nottingham

More information

Principal Components Analysis and Unsupervised Hebbian Learning

Principal Components Analysis and Unsupervised Hebbian Learning Princial Comonents Analysis and Unsuervised Hebbian Learning Robert Jacobs Deartment of Brain & Cognitive Sciences University of Rochester Rochester, NY 1467, USA August 8, 008 Reference: Much of the material

More information

Using a Computational Intelligence Hybrid Approach to Recognize the Faults of Variance Shifts for a Manufacturing Process

Using a Computational Intelligence Hybrid Approach to Recognize the Faults of Variance Shifts for a Manufacturing Process Journal of Industrial and Intelligent Information Vol. 4, No. 2, March 26 Using a Comutational Intelligence Hybrid Aroach to Recognize the Faults of Variance hifts for a Manufacturing Process Yuehjen E.

More information

Recursive Estimation of the Preisach Density function for a Smart Actuator

Recursive Estimation of the Preisach Density function for a Smart Actuator Recursive Estimation of the Preisach Density function for a Smart Actuator Ram V. Iyer Deartment of Mathematics and Statistics, Texas Tech University, Lubbock, TX 7949-142. ABSTRACT The Preisach oerator

More information

CHAPTER-II Control Charts for Fraction Nonconforming using m-of-m Runs Rules

CHAPTER-II Control Charts for Fraction Nonconforming using m-of-m Runs Rules CHAPTER-II Control Charts for Fraction Nonconforming using m-of-m Runs Rules. Introduction: The is widely used in industry to monitor the number of fraction nonconforming units. A nonconforming unit is

More information

A Recursive Block Incomplete Factorization. Preconditioner for Adaptive Filtering Problem

A Recursive Block Incomplete Factorization. Preconditioner for Adaptive Filtering Problem Alied Mathematical Sciences, Vol. 7, 03, no. 63, 3-3 HIKARI Ltd, www.m-hiari.com A Recursive Bloc Incomlete Factorization Preconditioner for Adative Filtering Problem Shazia Javed School of Mathematical

More information

Forced Oscillation Source Location via Multivariate Time Series Classification

Forced Oscillation Source Location via Multivariate Time Series Classification Forced Oscillation Source Location via ultivariate ime Series Classification Yao eng 1,2, Zhe Yu 1, Di Shi 1, Desong Bian 1, Zhiwei Wang 1 1 GEIRI North America, San Jose, CA 95134 2 North Carolina State

More information

One step ahead prediction using Fuzzy Boolean Neural Networks 1

One step ahead prediction using Fuzzy Boolean Neural Networks 1 One ste ahead rediction using Fuzzy Boolean eural etworks 1 José A. B. Tomé IESC-ID, IST Rua Alves Redol, 9 1000 Lisboa jose.tome@inesc-id.t João Paulo Carvalho IESC-ID, IST Rua Alves Redol, 9 1000 Lisboa

More information

E( x ) = [b(n) - a(n,m)x(m) ]

E( x ) = [b(n) - a(n,m)x(m) ] Exam #, EE5353, Fall 0. Here we consider MLPs with binary-valued inuts (0 or ). (a) If the MLP has inuts, what is the maximum degree D of its PBF model? (b) If the MLP has inuts, what is the maximum value

More information

Support Vector Machines versus Artificial Neural Networks for wood Dielectric Loss Factor estimation

Support Vector Machines versus Artificial Neural Networks for wood Dielectric Loss Factor estimation Suort Vector Machines versus Artificial Neural Networks for wood Dielectric Loss Factor estimation Lazaros Iliadis 1*, Stavros Tachos, Stavros Avramidis 3 Shawn Mansfield 3 1 Democritus University of Thrace,

More information

A numerical implementation of a predictor-corrector algorithm for sufcient linear complementarity problem

A numerical implementation of a predictor-corrector algorithm for sufcient linear complementarity problem A numerical imlementation of a redictor-corrector algorithm for sufcient linear comlementarity roblem BENTERKI DJAMEL University Ferhat Abbas of Setif-1 Faculty of science Laboratory of fundamental and

More information

Monopolist s mark-up and the elasticity of substitution

Monopolist s mark-up and the elasticity of substitution Croatian Oerational Research Review 377 CRORR 8(7), 377 39 Monoolist s mark-u and the elasticity of substitution Ilko Vrankić, Mira Kran, and Tomislav Herceg Deartment of Economic Theory, Faculty of Economics

More information

Neural network models for river flow forecasting

Neural network models for river flow forecasting Neural network models for river flow forecasting Nguyen Tan Danh Huynh Ngoc Phien* and Ashim Das Guta Asian Institute of Technology PO Box 4 KlongLuang 220 Pathumthani Thailand Abstract In this study back

More information

A Simple Fuzzy PI Control of Dual-Motor Driving Servo System

A Simple Fuzzy PI Control of Dual-Motor Driving Servo System MATEC Web of Conferences 04, 0006 (07) DOI: 0.05/ matecconf/07040006 IC4M & ICDES 07 A Simle Fuzzy PI Control of Dual-Motor Driving Servo System Haibo Zhao,,a, Chengguang Wang 3 Engineering Technology

More information

arxiv: v2 [quant-ph] 2 Aug 2012

arxiv: v2 [quant-ph] 2 Aug 2012 Qcomiler: quantum comilation with CSD method Y. G. Chen a, J. B. Wang a, a School of Physics, The University of Western Australia, Crawley WA 6009 arxiv:208.094v2 [quant-h] 2 Aug 202 Abstract In this aer,

More information

Generalized Coiflets: A New Family of Orthonormal Wavelets

Generalized Coiflets: A New Family of Orthonormal Wavelets Generalized Coiflets A New Family of Orthonormal Wavelets Dong Wei, Alan C Bovik, and Brian L Evans Laboratory for Image and Video Engineering Deartment of Electrical and Comuter Engineering The University

More information

A MIXED CONTROL CHART ADAPTED TO THE TRUNCATED LIFE TEST BASED ON THE WEIBULL DISTRIBUTION

A MIXED CONTROL CHART ADAPTED TO THE TRUNCATED LIFE TEST BASED ON THE WEIBULL DISTRIBUTION O P E R A T I O N S R E S E A R C H A N D D E C I S I O N S No. 27 DOI:.5277/ord73 Nasrullah KHAN Muhammad ASLAM 2 Kyung-Jun KIM 3 Chi-Hyuck JUN 4 A MIXED CONTROL CHART ADAPTED TO THE TRUNCATED LIFE TEST

More information

Robust Predictive Control of Input Constraints and Interference Suppression for Semi-Trailer System

Robust Predictive Control of Input Constraints and Interference Suppression for Semi-Trailer System Vol.7, No.7 (4),.37-38 htt://dx.doi.org/.457/ica.4.7.7.3 Robust Predictive Control of Inut Constraints and Interference Suression for Semi-Trailer System Zhao, Yang Electronic and Information Technology

More information

LINEAR SYSTEMS WITH POLYNOMIAL UNCERTAINTY STRUCTURE: STABILITY MARGINS AND CONTROL

LINEAR SYSTEMS WITH POLYNOMIAL UNCERTAINTY STRUCTURE: STABILITY MARGINS AND CONTROL LINEAR SYSTEMS WITH POLYNOMIAL UNCERTAINTY STRUCTURE: STABILITY MARGINS AND CONTROL Mohammad Bozorg Deatment of Mechanical Engineering University of Yazd P. O. Box 89195-741 Yazd Iran Fax: +98-351-750110

More information

Probability Estimates for Multi-class Classification by Pairwise Coupling

Probability Estimates for Multi-class Classification by Pairwise Coupling Probability Estimates for Multi-class Classification by Pairwise Couling Ting-Fan Wu Chih-Jen Lin Deartment of Comuter Science National Taiwan University Taiei 06, Taiwan Ruby C. Weng Deartment of Statistics

More information

Applied Mathematics and Computation

Applied Mathematics and Computation Alied Mathematics and Comutation 217 (2010) 1887 1895 Contents lists available at ScienceDirect Alied Mathematics and Comutation journal homeage: www.elsevier.com/locate/amc Derivative free two-oint methods

More information

Implementation of an Isotropic Elastic-Viscoplastic Model for Soft Soils using COMSOL Multiphysics

Implementation of an Isotropic Elastic-Viscoplastic Model for Soft Soils using COMSOL Multiphysics Imlementation of an Isotroic Elastic-Viscolastic Model for Soft Soils using COMSOL Multihysics M. Olsson 1,, T. Wood 1,, C. Alén 1 1 Division of GeoEngineering, Chalmers University of Technology, Gothenburg,

More information

Chapter 10. Supplemental Text Material

Chapter 10. Supplemental Text Material Chater 1. Sulemental Tet Material S1-1. The Covariance Matri of the Regression Coefficients In Section 1-3 of the tetbook, we show that the least squares estimator of β in the linear regression model y=

More information

A New Method of DDB Logical Structure Synthesis Using Distributed Tabu Search

A New Method of DDB Logical Structure Synthesis Using Distributed Tabu Search A New Method of DDB Logical Structure Synthesis Using Distributed Tabu Search Eduard Babkin and Margarita Karunina 2, National Research University Higher School of Economics Det of nformation Systems and

More information

Spacecraft Power System Controller Based on Neural Network

Spacecraft Power System Controller Based on Neural Network Proceedings of the 4 th International Middle East Power Systems Conference (MEPCON ), Cairo University, Egyt, December 9-2, 2, Paer ID 242. Sacecraft Power System Controller Based on Neural Network Hanaa

More information

Approximating min-max k-clustering

Approximating min-max k-clustering Aroximating min-max k-clustering Asaf Levin July 24, 2007 Abstract We consider the roblems of set artitioning into k clusters with minimum total cost and minimum of the maximum cost of a cluster. The cost

More information

Pairwise active appearance model and its application to echocardiography tracking

Pairwise active appearance model and its application to echocardiography tracking Pairwise active aearance model and its alication to echocardiograhy tracking S. Kevin Zhou 1, J. Shao 2, B. Georgescu 1, and D. Comaniciu 1 1 Integrated Data Systems, Siemens Cororate Research, Inc., Princeton,

More information

WAVELETS, PROPERTIES OF THE SCALAR FUNCTIONS

WAVELETS, PROPERTIES OF THE SCALAR FUNCTIONS U.P.B. Sci. Bull. Series A, Vol. 68, No. 4, 006 WAVELETS, PROPERTIES OF THE SCALAR FUNCTIONS C. PANĂ * Pentru a contrui o undină convenabilă Ψ este necesară şi suficientă o analiză multirezoluţie. Analiza

More information

Genetic Algorithm Based PID Optimization in Batch Process Control

Genetic Algorithm Based PID Optimization in Batch Process Control International Conference on Comuter Alications and Industrial Electronics (ICCAIE ) Genetic Algorithm Based PID Otimization in Batch Process Control.K. Tan Y.K. Chin H.J. Tham K.T.K. Teo odelling, Simulation

More information

Journal of Inequalities in Pure and Applied Mathematics

Journal of Inequalities in Pure and Applied Mathematics Journal of Inequalities in Pure and Alied Mathematics htt://jiam.vu.edu.au/ Volume 3, Issue 5, Article 8, 22 REVERSE CONVOLUTION INEQUALITIES AND APPLICATIONS TO INVERSE HEAT SOURCE PROBLEMS SABUROU SAITOH,

More information

Analysis of execution time for parallel algorithm to dertmine if it is worth the effort to code and debug in parallel

Analysis of execution time for parallel algorithm to dertmine if it is worth the effort to code and debug in parallel Performance Analysis Introduction Analysis of execution time for arallel algorithm to dertmine if it is worth the effort to code and debug in arallel Understanding barriers to high erformance and redict

More information

Iterative Methods for Designing Orthogonal and Biorthogonal Two-channel FIR Filter Banks with Regularities

Iterative Methods for Designing Orthogonal and Biorthogonal Two-channel FIR Filter Banks with Regularities R. Bregović and T. Saramäi, Iterative methods for designing orthogonal and biorthogonal two-channel FIR filter bans with regularities, Proc. Of Int. Worsho on Sectral Transforms and Logic Design for Future

More information

Various Proofs for the Decrease Monotonicity of the Schatten s Power Norm, Various Families of R n Norms and Some Open Problems

Various Proofs for the Decrease Monotonicity of the Schatten s Power Norm, Various Families of R n Norms and Some Open Problems Int. J. Oen Problems Comt. Math., Vol. 3, No. 2, June 2010 ISSN 1998-6262; Coyright c ICSRS Publication, 2010 www.i-csrs.org Various Proofs for the Decrease Monotonicity of the Schatten s Power Norm, Various

More information

EXACTLY PERIODIC SUBSPACE DECOMPOSITION BASED APPROACH FOR IDENTIFYING TANDEM REPEATS IN DNA SEQUENCES

EXACTLY PERIODIC SUBSPACE DECOMPOSITION BASED APPROACH FOR IDENTIFYING TANDEM REPEATS IN DNA SEQUENCES EXACTLY ERIODIC SUBSACE DECOMOSITION BASED AROACH FOR IDENTIFYING TANDEM REEATS IN DNA SEUENCES Ravi Guta, Divya Sarthi, Ankush Mittal, and Kuldi Singh Deartment of Electronics & Comuter Engineering, Indian

More information

Optimization of Gear Design and Manufacture. Vilmos SIMON *

Optimization of Gear Design and Manufacture. Vilmos SIMON * 7 International Conference on Mechanical and Mechatronics Engineering (ICMME 7) ISBN: 978--6595-44- timization of Gear Design and Manufacture Vilmos SIMN * Budaest Universit of Technolog and Economics,

More information

AR PROCESSES AND SOURCES CAN BE RECONSTRUCTED FROM. Radu Balan, Alexander Jourjine, Justinian Rosca. Siemens Corporation Research

AR PROCESSES AND SOURCES CAN BE RECONSTRUCTED FROM. Radu Balan, Alexander Jourjine, Justinian Rosca. Siemens Corporation Research AR PROCESSES AND SOURCES CAN BE RECONSTRUCTED FROM DEGENERATE MIXTURES Radu Balan, Alexander Jourjine, Justinian Rosca Siemens Cororation Research 7 College Road East Princeton, NJ 8 fradu,jourjine,roscag@scr.siemens.com

More information

CMSC 425: Lecture 4 Geometry and Geometric Programming

CMSC 425: Lecture 4 Geometry and Geometric Programming CMSC 425: Lecture 4 Geometry and Geometric Programming Geometry for Game Programming and Grahics: For the next few lectures, we will discuss some of the basic elements of geometry. There are many areas

More information

A TSK-Type Quantum Neural Fuzzy Network for Temperature Control

A TSK-Type Quantum Neural Fuzzy Network for Temperature Control International Mathematical Forum, 1, 2006, no. 18, 853-866 A TSK-Type Quantum Neural Fuzzy Network for Temperature Control Cheng-Jian Lin 1 Dept. of Computer Science and Information Engineering Chaoyang

More information