Variable Resolution Discretization in the Joint Space

Size: px
Start display at page:

Download "Variable Resolution Discretization in the Joint Space"

Transcription

1 Variable Reolution Dicretization in the Joint Space Chritopher K. Monon, David Wingate, and Kevin D. Seppi Computer Science, Brigham Young Univerity Todd S. Peteron Computer and Networking Science, Utah Valley State College Abtract We preent JoSTLe, an algorithm that perform value iteration on control problem with continuou action, allowing thi ueful reinforcement learning technique to be applied to problem where a priori action dicretization i inadequate. The algorithm i an extenion of a variable reolution technique that work for problem with continuou tate and dicrete action [6]. Reult are given that indicate that JoSTLe i a promiing tep toward reinforcement learning in a fully continuou domain. 1. Introduction Reinforcement Learning (RL) can be a ueful way of repreenting control problem becaue of it implicity. RL olution technique, uch a value iteration, can dicover complex olution that may be difficult to repreent or compute in cloed form. In the pecific cae of value iteration, a problem i olved by computing an approximation to the value function. The value function may then be ued to create a control policy. [4] One of the more attractive propertie of value iteration i that it may be performed iteratively. The value of each point in pace may be updated according to an equation like the following: V (, a)= τ γ t R((t), a) dt + γ τ up V ((τ), a ) (1) a where V (, a) denote the value of the initial tate and action a, (t) i the tate reulting from the application of a for t unit of time, R((t), a) i the current reinforcement, γ [, 1) i the dicount factor, and τ i the amount of time for which a i executed. Note that thi aume a determinitic environment, o tranition probabilitie have been omitted. 1 1 Additionally, even though thi definition of the value function look like a quality function, we maintain the nomenclature of value iteration o a to avoid confuion with Q-learning. Implementing (1) directly i impoible. In a dicrete etting, (1) become omething like the following: T 1 V (, a)=φ γ tφ R((tφ), a) + γ T φ up V ((T φ), a ) t= a (2) where φ i a problem-pecific time tep. With dicrete action, the up operator become a max operator and a imple linear earch i ufficient to implement it. In domain with continuou action, however, the preence of the upremum i problematic, a a perfect implementation would require an exhautive earch of an infinite pace. Fortunately thi iue doe not plague all continuou action RL problem; a well-known reult from optimal control theory tate that many minimum time control problem may be olved optimally uing only bang-bang control [2]. Thi fact allow many reearcher to optimally dicretize continuou action pace a priori. Though many intereting problem may be olved uing a naive dicretization, many other may not. Even if a problem may be olved uing bang-bang control in imulation, the policy generated can rarely be run on real hardware. A method for olving thee problem without a tatic action dicretization i needed. Thi paper preent the Joint Space Triangulation Learner (JoSTLe), which enable value iteration to olve problem with continuou action. It i baed on Variable Reolution Dicretization (VRD) a preented by Muño and Moore [6], and relie on the ame fundamental obervation that not all portion of the problem pace are of equal importance. JoSTLe ue a homogeneou data tructure to dynamically allocate reource acro both tate and action pace. In the ame way that VRD allow each problem to dictate region of interet in the tate pace, JoSTLe allow each problem to dictate thoe region in the combined tateaction (or joint ) pace. Thi create the poibility for dicretizing action differently at each tate. Thi i not the firt work to addre continuou action problem, but it i to our knowledge the firt to work with general continuou tate and action problem. Other ap-

2 v = (,a) Action Figure 1. The Kuhn Triangulation of a cube proache typically involve uing one dicretization technique on the tate pace and then performing ome form of regreion in the action pace [1] at each tate. Thi approach i ueful, but can run into problem when the value function ha dicontinuou boundarie. It can alo be inefficient when one action repreentation could pan multiple tate. 2. Baic Variable Reolution Dicretization VRD dicretize the d -dimenional tate pace into hypercube, arranged hierarchically in a kd-trie. The root node cover the entire pace. At every branch, a plit i performed in one of the tate dimenion, creating two maller hypercube. A Kuhn triangulation i implemented at each leaf, effectively plitting it into d! implice (Figure 1). The overall effect i a complete triangulation of the pace. The value function i interpolated linearly within thi triangulation uing barycentric coordinate [6]. VRD proceed in two phae: value iteration and tate pace refinement. It begin with a rough initial dicretization of the tate pace. The dicretization i ued to perform value iteration until it converge. The information gained from the value iteration proce i ued to refine the dicretization, ideally plitting in area that require finer repreentation to generate a good policy. The iteration and refinement tep are repeated until a atifactory repreentation of the value function i obtained. The kd-trie allow for efficient point localization while the Kuhn triangulation allow for efficient interpolation. A full decription of the algorithm with comprehenive citation on component element i contained in [6]. 3. Extending to the Joint Space JoSTLe i an extenion of VRD but retain many of it characteritic, including the kd-trie and Kuhn triangulation. The primary difference between the two i that JoS- TLe work in the joint tate-action pace rather than jut the tate pace. Figure 2. Final tate exit adjacent hypercube (haded) Let d be the dimenionality of the tate pace and d a be the dimenionality of the action pace. We may then define the joint pace a the Carteian product of tate and action pace, yielding a pace whoe dimenionality i given by d = d + d a. Thi joint pace i teellated by hypercube in the ame manner a VRD tate pace. Each vertex of each cube i compoed of the concatenation of tate and action vector: v = (, a). Aociated with each vertex i a value V (, a), ometime abbreviated a V (v). Thi extenion into a higher dimenional pace repreent the primary mathematical difference between JoSTLe and VRD. Equation (2) till applie directly. There i another minor difference in the way that a trajectory topping point i determined. In VRD a trajectory top when it exit the initial implex. JoSTLe top once the tate no longer interect any hypercube adjacent to the initial tate/action point. Figure 2 illutrate thi idea. The circled vertex i the tarting point, and the trajectory in the tate pace doe not end until it i no longer interecting any of that vertex hypercube. Thi paper addree the iue of implementing uch a ytem. The addition of action dimenion to the dicretization poe ome unique problem that require attention. The firt iue i that of finding action at each tate that produce the maximum value. The econd iue i related to the generalization of the earch algorithm to arbitrary dimenion. The third i that of deciding when and how to plit hypercube. Thee iue are addreed in the next ection Searching for Maxima The earch for action that produce the maximum value at a given tate i a fundamental part of value iteration. In traditional value iteration the action pace i dicrete and homogeneou. In the dicretization produced by JoS- TLe the action pace i continuou with a heterogeneou dicretization. The piecewie linear interpolation imple-

3 a 1 a Figure 3. A joint pace implex mented in JoSTLe addree thi problem effectively. Figure 3 illutrate a joint pace with two tate dimenion and one action dimenion. The teellation generated by the kd-trie would actually cover the entire pace with hypercube, each of which would be triangulated, but for purpoe of explanation only one triangle i hown. The action earch problem may be viewed a a earch along the line in the figure. The line repreent a region of contant tate and variable action. Region of thi nature mut be earched at every tep of value iteration in order to calculate a dicounted value. While in general thi i a nonlinear programming problem, the repreentation allow for ignificant implification. Becaue the interior value of each implex follow a linear function, the maximum mut occur at a implex boundary. It cannot occur uniquely at the interior. Thi i eaily proven by howing that the gradient of the interpolation function i contant. The proof that thi hold even when contrained along a region uch a that in Figure 3 i alo fairly traightforward [5]. Thi inight allow u to retrict our earch to the point where the line (a depicted in Figure 3) interect with the implex boundarie, effectively tranforming a continuou problem into a dicrete problem. The earch i performed by finding the interection point and picking the maximum Generalization to Arbitrary Dimenion Finding the interection point can be difficult. Figure 4 illutrate the problem and an inight that erve to generalize the algorithm. Each poible patial interection generate a different kind of earch pace. In Figure 4(a) the region of interet are point on a line. In Figure 4(b), however, the intereting portion of the pace are vertice of a triangle formed by licing a tetrahedron with a plane. In general, we are alway eeking the vertice of a implex formed from uch a lice, and the hape of that implex will change baed on the dimenionalitie of the tate and action pace. Linear interpolation once again allow u to take a ueful hortcut. Figure 4(a) and 4(b) alo how the projection of 1 a2 (a) d = 2, d a = 1 (b) d = 1, d = 2 a1 Figure 4. Joint pace projection the implex into the tate pace. Note that in both projection, the region of interet i a ingle point in overlapping implice. Again, becaue interpolation i linear, the point of interection are eaily found by projecting the boundarie onto the tate pace and performing interpolation on thee lower dimenional hadow. Though boundarie will overlap in the lower dimenional pace, each will have a unique et of vertice and produce different anwer. One of thee will be the maximum. The algorithm for earching the action pace at a given tate thu become very imple and elegant: Find all hypercube interected by the hyperplane at. Thi i eaily done with a kd-trie uing an orthogonal range query. For each hypercube, enumerate all Kuhn implice. For each implex, boundarie. enumerate all d -dimenional Project each boundary into the tate pace and interpolate at. The proof that thi method i equivalent to finding interection in the joint pace i given in [5] Splitting Criteria In principle, it would be bet to refine the joint pace only where it will improve the policy. Unfortunately, thi i

4 not generally computable in advance. An alternative would be to refine region of the pace that improve the value function etimate. Thi i alo not generally knowable, but an approximation may be made. We refine the model baed on maximum interpolation error. The error function i defined over all point p = (, a) within each implex S a E(p) = V (p) V (p) (3) where V (p) i the multitep dicounted value at p, and V (p) i the interpolated value at p, which i a weighted um of the V (p) value of the vertice in the encloing implex. Thi error function i defined everywhere in the interior of each hypercube. The purpoe of plitting i to reduce the maximum error. Splitting i therefore done when up p E(p) urpae ome uitable threhold ɛ. In practice, up p E(p) i approximated with a et of random ample point P H within a hypercube by max p PH E(p). Whenever thi value i greater than ɛ for a hypercube, the cube i marked for plitting. 2 Once all cube have been evaluated, the marked cube are plit. What remain i to determine in which dimenion to plit them. We plit in the dimenion that produce a new cube with the mallet maximum error. More preciely, let C L x (H) and CR x (H) be the left and right children reulting from a plit of hypercube H in dimenion x: [ ( )] x = arg min x {,,d} min D {L,R} max p P C D x E(p). (4) Thi may reult in two child cube with very different error. The plit i performed o a to minimize the maximum error of one of the cube, leaving the poibility that the other will have a very high error. Thi i tolerable becaue the high error cube i likely to be plit at the next iteration to improve it error characteritic. No plitting i done if the hypercube ha low error or if the hypercube i maller than the mallet feature of interet, a parameter decribed in the next ection JoSTLe Parameter JoSTLe add a mall number of tunable parameter to the tandard value iteration algorithm, hown in Table 1. The minimum feature length in dimenion i i denoted ω i. Thi parameter determine when a cube i too mall to be plit regardle of error and it erve to keep the algorithm from plitting forever around dicontinuou boundarie in the value function. Thi parameter i typically eay 2 Random ampling i a naive and imple way to approach the problem, but it i not likely to be the bet. Other approximation are the ubject of future work. ω i Ω ɛ σ Table 1. JoSTLe parameter Minimum feature length in dimenion i Minimum Lebegue meaure of a hypercube Error threhold Number of ample point per hypercube to obtain given that many reinforcement learning problem have known reward boundarie. The mallet reward feature ize i often a good tarting point for thi parameter. In practice, the performance of the algorithm degrade moothly a thi parameter i increaed (a plit are limited). Smaller value alway yield better accuracy, but often at the expene of convergence peed. The ω i parameter may be ued in a number of way. It can limit a cube ability to plit in a particular dimenion if it length in that dimenion i le than ω i. Alternatively, it can be ued to compute a minimum allowed Lebegue meaure Ω = d i=1 ω i. If thi latter method i ued, then a cube i not plit in any dimenion if it Lebegue meaure i maller than Ω. The experiment outlined in thi paper ue the former, though the latter wa teted with imilar reult. The error threhold ɛ i alo ued during the plitting proce. If a cube maximum error i le than ɛ, the cube need not be plit. Determining an appropriate value for ɛ can be challenging, but a practical approximation may be made baed on the maximum range of reward value R max R min, the time tep φ, and the dicount factor γ. If the problem ha only terminal reinforcement, an upper bound on the error i given by φ(r max R min ) ince the integral over trajectory reward will be zero until the end of the trajectory. Although thi i a fairly conervative bound, it work well in practice. Problem with non-terminal reinforcement may ue an alternate upper bound, determined by accounting for an infinite tring of dicounted reward: φ(rmax Rmin) 1 γ. Thee upper bound can be ued to define a more intuitive error threhold. For example, the threhold may be et to ome fraction of the upper bound, making it eay to generate reaonable value. 1% of the maximum error i often a reaonable error threhold. The number of point σ cattered in a given hypercube may be computed from the minimum Lebegue meaure: σ = ΩH where Ω H i the Lebegue meaure of hypercube H. The mallet allowable feature i allocated exactly one point. If σ become 1, then the hypercube can no longer be effectively teted for plitting. Initially, the number of ample point can be very large, reulting in a ubtantial increae in time pent ampling Ω (5)

5 1 5 Action State Figure 5. 1D Golf reward boundarie (a) Dicretization (b) Policy Figure 6. JoSTLe dicretization and policy for 1D Golf and teting for plit. Thi number may be limited to a reaonable maximum, e.g Convergence Gordon addreed the iue of convergence at length, and howed that averaging function approximator will allow the value iteration proce to converge [3]. Among thee are barycentric interpolator, of which the linear interpolation method decribed here i one. Becaue value iteration i done eparately from the refinement proce, and we operate over a finite et of action, Gordon convergence reult till applie. 4. Experiment D Golf One-dimenional Golf i a tet problem with low dimenionality and a continuou tate and action pace. A golf ball i itting on a one-dimenional line and mut be hit into a hole in the center of the pace. The tate pace i decribed by [ 1, 1]. The action pace i alo continuou: a [ 1, 1]. The hole i centered at (, ) and i.5 unit wide. The environment i determinitic and acceible. The ytem characteritic are t+1 = t + a 1 a. (6) a If the ball hit a wall, it top and a reinforcement of 1 i received. If it land in the hole (i.e., t+1 [.25,.25]) a reinforcement of 1 i received. In all other cae, a reinforcement of i received. A graphical repreentation of the joint pace with the poitive (in the center) and negative (in the corner) reward boundarie i hown in Figure 5. Thi problem i more intereting than it firt appear. The region of high reward i very mall and nonlinear. Additionally, reinforcement are not located trictly at the boundary of the problem pace, making them difficult for VRD to find Reult Both JoSTLe and VRD were applied to the golf problem. Though there are many poible plitting criteria for VRD, in a 1-dimenional problem average corner value difference work a well a any of them (more complex criteria are only helpful in higher dimenion [6]). In VRD, plitting occurred if the value difference wa above.1, and in the cae of JoSTLe, ɛ wa et to 5% of the upper bound on the error. The timetep φ = 1. Both ued a γ of ince it wa known a priori that only one tep i ever needed. JoSTLe began with a ingle joint pace hypercube and learned the appropriate dicretization over time. VRD began with a ingle line egment in the tate pace and wa applied uing everal different uniform action dicretization. For each algorithm, policy accuracy wa calculated after every round of plitting and iteration. Since the optimal policy i known for thi problem and alway conit of a ingle tep, the accuracy wa calculated by canning the tate pace and querying the model for correct policy value. The accuracy i the ratio of correct action to total tate queried. The policy obtained by the joint learner i hown in Figure 6. In all available tate, a correct action i choen (the accuracy i 1%). The dicretization hypercube are alo hown in Figure 6; it i clear that the learner concentrated it reource on area of harp reward tranition. Thi behavior i expected ince γ =.

6 Accuracy Joint MM 2.1 MM 16 MM Vertice (Log Scale) Figure 7. JoSTLe and VRD accuracy v. number of vertice Velocity Poition Figure 8. JoSTLe policy for Mountain Car The accuracy of JoSTLe v. the number of vertice ued i hown in Figure 7. The behavior of VRD i hown on the ame graph. The numbered label indicate the number of dicrete action available to the algorithm throughout it trial. The graph how that JoSTLe policy accuracy went up quickly with every refinement, while the accuracy obtained with VRD roe lowly. It took 256 available action to compare to the accuracy of JoSTLe, and far more tate/action pair. Given fewer action, VRD peak at a particular policy accuracy and then level off, ince finer action dicretization i required but not available Additional Experiment The firt additional experiment wa alo done uing the 1D Golf problem, altered o that the reward boundarie did not cover the entire tate pace. Thi required JoSTLe to do real value iteration (with a nonzero dicount factor). The reult were jut a good a with the original problem. Additionally, JoSTLe found a nearly optimal policy for the Mountain Car problem [6] without any prior knowledge either of what action would be ueful or of which part of the tate pace were intereting. It learned that the two bang-bang action are more ueful than the other (full forward and full revere). Thee turn out to be exactly the ame two action ued in [6]. The policy learned by JoSTLe i hown in Figure 8. The policy i not perfect, but i cloe to optimal. Some artifact exit and the reaon for their exitence i till being explored. Alternate view of the policy are hown in Figure 9. Figure 9(a) how the full policy and highlight the fact that JoSTLe focued mot of it attention on two action: full forward and full revere. Thee are the bang-bang action ued by VRD. Figure 9(b) and 9(c) how the pol- icy projected onto the Poition/Acceleration and Velocity/Acceleration plane. The preence of point at the interior of thee figure, rather than excluively at the top and bottom, indicate that other uboptimal action crept into the policy in ome area of the tate pace, a matter which need to be tudied further. 5. Liabilitie One unfortunate characteritic of JoSTLe when compared to VRD i it higher complexity. Becaue the dimenionality i increaed, all of the wort pace and time characteritic of VRD are exacerbated in JoSTLe (e.g. d i higher for JoSTLe, and each plit till produce 2 d new vertice). Additionally, while VRD never enumerate the implice of a hypercube, JoSTLe mut. Each hypercube ha exactly d! Kuhn implice, each of which JoSTLe mut decompoe into all ( ) d d boundarie. Thi ha a negative impact on dimenional calability. Some optimization can alleviate the complexity, placing it on a more equal footing with VRD. The culling of degenerate and redundant implice, a well a the fact that JoSTLe often need fewer node overall can help to ignificantly reduce the complexity in practice. More work mut be done to determine where ele the complexity may be reduced. Another problem i expoed by the fact that a perfectly optimal policy for Mountain Car wa never achieved. Reearch revealed everal potential area of improvement. Firt, it i not clear that the plitting criterion preented here i exactly what i needed to generate a good policy. Second, though the integrity of the final value update equation wa maintained throughout the development of the algorithm, it appear that one of the baic aumption of value iteration may have been violated: the Markov property. VRD mucled itelf into retaining thi property

7 in a continuou pace by treating interpolation weight a tranition probabilitie. JoSTLe ha no imilar interpretation of interpolated point, a it mut firt chooe an action to determine the mot likely current tate. In other word, we don t know where we are until we go omewhere ele, a clear violation of the requirement that our deciion be baed only on the current tate. That JoSTLe work a well a it doe indicate that the violation may not be eriou, but it doe merit further exploration and hould be addreed in a future work. 6. Concluion and Future Reearch Acceleration Acceleration (a) 3D Policy Poition (b) Po/Accel projection Velocity (c) Vel/Accel projection Figure 9. JoSTLe policy for Mountain Car JoSTLe repreent a productive tep toward the ability to perform value iteration on problem with continuou action. It provide a homogeneou framework for refinement of both tate and action and ha an elegant appeal. Even o, there i much room for improvement. The Markov property need to be tudied in greater detail in thi context to determine whether there i a ueful interpretation of JoSTLe that doe not violate thi property. Reearch in that area i ongoing. Additionally, VRD propoed influence and variance to dicretize only thoe portion of the pace that affect the overall policy; more reearch could be devoted to an analyi of joint-pace analogue. Non-uniform plitting i another poible reearch direction. Both JoSTLe and VRD plit region of pace in half, which i not alway optimal in term of efficiency. The ability to perform oblique plit may alo allow for a more efficient repreentation. Some preliminary work in thi area indicate promie, but more reearch i needed. Reference [1] L. Baird and A. Klopf. Reinforcement learning with highdimenional, continuou action. Technical Report WL-TR , Wright-Patteron Air Force Bae, Ohio, [2] B. Friedland. In Advanced Control Sytem Deign, New Jerey, Prentice Hall. [3] G. J. Gordon. Stable function approximation in dynamic programming. In A. Priediti and S. Ruell, editor, Proceeding of the Twelfth International Conference on Machine Learning, page , San Francico, CA, Morgan Kaufmann. [4] L. P. Kaelbling, M. L. Littman, and A. P. Moore. Reinforcement learning: A urvey. Journal of Artificial Intelligence Reearch, 4: , [5] C. K. Monon. Reinforcement learning in the joint pace: Value iteration in world with continuou tate and action. Mater thei, Brigham Young Univerity, Computer Science Department, Apr. 23. [6] R. Muño and A. Moore. Variable reolution dicretization in optimal control. Machine Learning, 49, Number 2/3: , November/December 22.

Variable Resolution Discretization in the Joint Space

Variable Resolution Discretization in the Joint Space Brigham Young University BYU ScholarsArchive All Faculty Publications 2004-12-18 Variable Resolution Discretization in the Joint Space Christopher K. Monson shiblon@gmail.com Kevin Seppi Brigham Young

More information

Universität Augsburg. Institut für Informatik. Approximating Optimal Visual Sensor Placement. E. Hörster, R. Lienhart.

Universität Augsburg. Institut für Informatik. Approximating Optimal Visual Sensor Placement. E. Hörster, R. Lienhart. Univerität Augburg à ÊÇÅÍÆ ËÀǼ Approximating Optimal Viual Senor Placement E. Hörter, R. Lienhart Report 2006-01 Januar 2006 Intitut für Informatik D-86135 Augburg Copyright c E. Hörter, R. Lienhart Intitut

More information

Performance of a Robust Filter-based Approach for Contour Detection in Wireless Sensor Networks

Performance of a Robust Filter-based Approach for Contour Detection in Wireless Sensor Networks Performance of a Robut Filter-baed Approach for Contour Detection in Wirele Senor Network Hadi Alati, William A. Armtrong, Jr., and Ai Naipuri Department of Electrical and Computer Engineering The Univerity

More information

Topics. Lecture 37: Global Optimization. Issues. A Simple Example: Copy Propagation X := 3 B > 0 Y := 0 X := 4 Y := Z + W A := 2 * 3X

Topics. Lecture 37: Global Optimization. Issues. A Simple Example: Copy Propagation X := 3 B > 0 Y := 0 X := 4 Y := Z + W A := 2 * 3X Lecture 37: Global Optimization [Adapted from note by R. Bodik and G. Necula] Topic Global optimization refer to program optimization that encompa multiple baic block in a function. (I have ued the term

More information

Quadrilaterals. Learning Objectives. Pre-Activity

Quadrilaterals. Learning Objectives. Pre-Activity Section 3.4 Pre-Activity Preparation Quadrilateral Intereting geometric hape and pattern are all around u when we tart looking for them. Examine a row of fencing or the tiling deign at the wimming pool.

More information

xy-monotone path existence queries in a rectilinear environment

xy-monotone path existence queries in a rectilinear environment CCCG 2012, Charlottetown, P.E.I., Augut 8 10, 2012 xy-monotone path exitence querie in a rectilinear environment Gregory Bint Anil Mahehwari Michiel Smid Abtract Given a planar environment coniting of

More information

Drawing Lines in 2 Dimensions

Drawing Lines in 2 Dimensions Drawing Line in 2 Dimenion Drawing a traight line (or an arc) between two end point when one i limited to dicrete pixel require a bit of thought. Conider the following line uperimpoed on a 2 dimenional

More information

MAT 155: Describing, Exploring, and Comparing Data Page 1 of NotesCh2-3.doc

MAT 155: Describing, Exploring, and Comparing Data Page 1 of NotesCh2-3.doc MAT 155: Decribing, Exploring, and Comparing Data Page 1 of 8 001-oteCh-3.doc ote for Chapter Summarizing and Graphing Data Chapter 3 Decribing, Exploring, and Comparing Data Frequency Ditribution, Graphic

More information

An Intro to LP and the Simplex Algorithm. Primal Simplex

An Intro to LP and the Simplex Algorithm. Primal Simplex An Intro to LP and the Simplex Algorithm Primal Simplex Linear programming i contrained minimization of a linear objective over a olution pace defined by linear contraint: min cx Ax b l x u A i an m n

More information

Minimum congestion spanning trees in bipartite and random graphs

Minimum congestion spanning trees in bipartite and random graphs Minimum congetion panning tree in bipartite and random graph M.I. Otrovkii Department of Mathematic and Computer Science St. John Univerity 8000 Utopia Parkway Queen, NY 11439, USA e-mail: otrovm@tjohn.edu

More information

Chapter 13 Non Sampling Errors

Chapter 13 Non Sampling Errors Chapter 13 Non Sampling Error It i a general aumption in the ampling theory that the true value of each unit in the population can be obtained and tabulated without any error. In practice, thi aumption

More information

Delaunay Triangulation: Incremental Construction

Delaunay Triangulation: Incremental Construction Chapter 6 Delaunay Triangulation: Incremental Contruction In the lat lecture, we have learned about the Lawon ip algorithm that compute a Delaunay triangulation of a given n-point et P R 2 with O(n 2 )

More information

Modeling of underwater vehicle s dynamics

Modeling of underwater vehicle s dynamics Proceeding of the 11th WEA International Conference on YTEM, Agio Nikolao, Crete Iland, Greece, July 23-25, 2007 44 Modeling of underwater vehicle dynamic ANDRZEJ ZAK Department of Radiolocation and Hydrolocation

More information

A SIMPLE IMPERATIVE LANGUAGE THE STORE FUNCTION NON-TERMINATING COMMANDS

A SIMPLE IMPERATIVE LANGUAGE THE STORE FUNCTION NON-TERMINATING COMMANDS A SIMPLE IMPERATIVE LANGUAGE Eventually we will preent the emantic of a full-blown language, with declaration, type and looping. However, there are many complication, o we will build up lowly. Our firt

More information

Motion Control (wheeled robots)

Motion Control (wheeled robots) 3 Motion Control (wheeled robot) Requirement for Motion Control Kinematic / dynamic model of the robot Model of the interaction between the wheel and the ground Definition of required motion -> peed control,

More information

Kinematics Programming for Cooperating Robotic Systems

Kinematics Programming for Cooperating Robotic Systems Kinematic Programming for Cooperating Robotic Sytem Critiane P. Tonetto, Carlo R. Rocha, Henrique Sima, Altamir Dia Federal Univerity of Santa Catarina, Mechanical Engineering Department, P.O. Box 476,

More information

CENTER-POINT MODEL OF DEFORMABLE SURFACE

CENTER-POINT MODEL OF DEFORMABLE SURFACE CENTER-POINT MODEL OF DEFORMABLE SURFACE Piotr M. Szczypinki Iintitute of Electronic, Technical Univerity of Lodz, Poland Abtract: Key word: Center-point model of deformable urface for egmentation of 3D

More information

Lecture 14: Minimum Spanning Tree I

Lecture 14: Minimum Spanning Tree I COMPSCI 0: Deign and Analyi of Algorithm October 4, 07 Lecture 4: Minimum Spanning Tree I Lecturer: Rong Ge Scribe: Fred Zhang Overview Thi lecture we finih our dicuion of the hortet path problem and introduce

More information

Computer Arithmetic Homework Solutions. 1 An adder for graphics. 2 Partitioned adder. 3 HDL implementation of a partitioned adder

Computer Arithmetic Homework Solutions. 1 An adder for graphics. 2 Partitioned adder. 3 HDL implementation of a partitioned adder Computer Arithmetic Homework 3 2016 2017 Solution 1 An adder for graphic In a normal ripple carry addition of two poitive number, the carry i the ignal for a reult exceeding the maximum. We ue thi ignal

More information

SLA Adaptation for Service Overlay Networks

SLA Adaptation for Service Overlay Networks SLA Adaptation for Service Overlay Network Con Tran 1, Zbigniew Dziong 1, and Michal Pióro 2 1 Department of Electrical Engineering, École de Technologie Supérieure, Univerity of Quebec, Montréal, Canada

More information

The Association of System Performance Professionals

The Association of System Performance Professionals The Aociation of Sytem Performance Profeional The Computer Meaurement Group, commonly called CMG, i a not for profit, worldwide organization of data proceing profeional committed to the meaurement and

More information

CSE 250B Assignment 4 Report

CSE 250B Assignment 4 Report CSE 250B Aignment 4 Report March 24, 2012 Yuncong Chen yuncong@c.ucd.edu Pengfei Chen pec008@ucd.edu Yang Liu yal060@c.ucd.edu Abtract In thi project, we implemented the recurive autoencoder (RAE) a decribed

More information

3D SMAP Algorithm. April 11, 2012

3D SMAP Algorithm. April 11, 2012 3D SMAP Algorithm April 11, 2012 Baed on the original SMAP paper [1]. Thi report extend the tructure of MSRF into 3D. The prior ditribution i modified to atify the MRF property. In addition, an iterative

More information

LinkGuide: Towards a Better Collection of Hyperlinks in a Website Homepage

LinkGuide: Towards a Better Collection of Hyperlinks in a Website Homepage Proceeding of the World Congre on Engineering 2007 Vol I LinkGuide: Toward a Better Collection of Hyperlink in a Webite Homepage A. Ammari and V. Zharkova chool of Informatic, Univerity of Bradford anammari@bradford.ac.uk,

More information

UC Berkeley International Conference on GIScience Short Paper Proceedings

UC Berkeley International Conference on GIScience Short Paper Proceedings UC Berkeley International Conference on GIScience Short Paper Proceeding Title A novel method for probabilitic coverage etimation of enor network baed on 3D vector repreentation in complex urban environment

More information

Planning of scooping position and approach path for loading operation by wheel loader

Planning of scooping position and approach path for loading operation by wheel loader 22 nd International Sympoium on Automation and Robotic in Contruction ISARC 25 - September 11-14, 25, Ferrara (Italy) 1 Planning of cooping poition and approach path for loading operation by wheel loader

More information

Stochastic Search and Graph Techniques for MCM Path Planning Christine D. Piatko, Christopher P. Diehl, Paul McNamee, Cheryl Resch and I-Jeng Wang

Stochastic Search and Graph Techniques for MCM Path Planning Christine D. Piatko, Christopher P. Diehl, Paul McNamee, Cheryl Resch and I-Jeng Wang Stochatic Search and Graph Technique for MCM Path Planning Chritine D. Piatko, Chritopher P. Diehl, Paul McNamee, Cheryl Rech and I-Jeng Wang The John Hopkin Univerity Applied Phyic Laboratory, Laurel,

More information

Advanced Encryption Standard and Modes of Operation

Advanced Encryption Standard and Modes of Operation Advanced Encryption Standard and Mode of Operation G. Bertoni L. Breveglieri Foundation of Cryptography - AES pp. 1 / 50 AES Advanced Encryption Standard (AES) i a ymmetric cryptographic algorithm AES

More information

1 The secretary problem

1 The secretary problem Thi i new material: if you ee error, pleae email jtyu at tanford dot edu 1 The ecretary problem We will tart by analyzing the expected runtime of an algorithm, a you will be expected to do on your homework.

More information

Cutting Stock by Iterated Matching. Andreas Fritsch, Oliver Vornberger. University of Osnabruck. D Osnabruck.

Cutting Stock by Iterated Matching. Andreas Fritsch, Oliver Vornberger. University of Osnabruck. D Osnabruck. Cutting Stock by Iterated Matching Andrea Fritch, Oliver Vornberger Univerity of Onabruck Dept of Math/Computer Science D-4909 Onabruck andy@informatikuni-onabrueckde Abtract The combinatorial optimization

More information

A System Dynamics Model for Transient Availability Modeling of Repairable Redundant Systems

A System Dynamics Model for Transient Availability Modeling of Repairable Redundant Systems International Journal of Performability Engineering Vol., No. 3, May 05, pp. 03-. RAMS Conultant Printed in India A Sytem Dynamic Model for Tranient Availability Modeling of Repairable Redundant Sytem

More information

Motivation: Level Sets. Input Data Noisy. Easy Case Use Marching Cubes. Intensity Varies. Non-uniform Exposure. Roger Crawfis

Motivation: Level Sets. Input Data Noisy. Easy Case Use Marching Cubes. Intensity Varies. Non-uniform Exposure. Roger Crawfis Level Set Motivation: Roger Crawfi Slide collected from: Fan Ding, Charle Dyer, Donald Tanguay and Roger Crawfi 4/24/2003 R. Crawfi, Ohio State Univ. 109 Eay Cae Ue Marching Cube Input Data Noiy 4/24/2003

More information

Gray-level histogram. Intensity (grey-level) transformation, or mapping. Use of intensity transformations:

Gray-level histogram. Intensity (grey-level) transformation, or mapping. Use of intensity transformations: Faculty of Informatic Eötvö Loránd Univerity Budapet, Hungary Lecture : Intenity Tranformation Image enhancement by point proceing Spatial domain and frequency domain method Baic Algorithm for Digital

More information

Generic Traverse. CS 362, Lecture 19. DFS and BFS. Today s Outline

Generic Traverse. CS 362, Lecture 19. DFS and BFS. Today s Outline Generic Travere CS 62, Lecture 9 Jared Saia Univerity of New Mexico Travere(){ put (nil,) in bag; while (the bag i not empty){ take ome edge (p,v) from the bag if (v i unmarked) mark v; parent(v) = p;

More information

Development of an atmospheric climate model with self-adapting grid and physics

Development of an atmospheric climate model with self-adapting grid and physics Intitute of Phyic Publihing Journal of Phyic: Conference Serie 16 (2005) 353 357 doi:10.1088/1742-6596/16/1/049 SciDAC 2005 Development of an atmopheric climate model with elf-adapting grid and phyic Joyce

More information

Routing Definition 4.1

Routing Definition 4.1 4 Routing So far, we have only looked at network without dealing with the iue of how to end information in them from one node to another The problem of ending information in a network i known a routing

More information

Edits in Xylia Validity Preserving Editing of XML Documents

Edits in Xylia Validity Preserving Editing of XML Documents dit in Xylia Validity Preerving diting of XML Document Pouria Shaker, Theodore S. Norvell, and Denni K. Peter Faculty of ngineering and Applied Science, Memorial Univerity of Newfoundland, St. John, NFLD,

More information

IMPLEMENTATION OF AREA, VOLUME AND LINE SOURCES

IMPLEMENTATION OF AREA, VOLUME AND LINE SOURCES December 01 ADMS 5 P503I1 IMPEMENTATION OF AREA, VOUME AND INE SOURCES The Met. Office (D J Thomon) and CERC 1. INTRODUCTION ADMS model line ource, and area and volume ource with conve polgon bae area.

More information

Hassan Ghaziri AUB, OSB Beirut, Lebanon Key words Competitive self-organizing maps, Meta-heuristics, Vehicle routing problem,

Hassan Ghaziri AUB, OSB Beirut, Lebanon Key words Competitive self-organizing maps, Meta-heuristics, Vehicle routing problem, COMPETITIVE PROBABIISTIC SEF-ORGANIZING MAPS FOR ROUTING PROBEMS Haan Ghaziri AUB, OSB Beirut, ebanon ghaziri@aub.edu.lb Abtract In thi paper, we have applied the concept of the elf-organizing map (SOM)

More information

Markov Random Fields in Image Segmentation

Markov Random Fields in Image Segmentation Preented at SSIP 2011, Szeged, Hungary Markov Random Field in Image Segmentation Zoltan Kato Image Proceing & Computer Graphic Dept. Univerity of Szeged Hungary Zoltan Kato: Markov Random Field in Image

More information

else end while End References

else end while End References 621-630. [RM89] [SK76] Roenfeld, A. and Melter, R. A., Digital geometry, The Mathematical Intelligencer, vol. 11, No. 3, 1989, pp. 69-72. Sklanky, J. and Kibler, D. F., A theory of nonuniformly digitized

More information

Representations and Transformations. Objectives

Representations and Transformations. Objectives Repreentation and Tranformation Objective Derive homogeneou coordinate tranformation matrice Introduce tandard tranformation - Rotation - Tranlation - Scaling - Shear Scalar, Point, Vector Three baic element

More information

Operational Semantics Class notes for a lecture given by Mooly Sagiv Tel Aviv University 24/5/2007 By Roy Ganor and Uri Juhasz

Operational Semantics Class notes for a lecture given by Mooly Sagiv Tel Aviv University 24/5/2007 By Roy Ganor and Uri Juhasz Operational emantic Page Operational emantic Cla note for a lecture given by Mooly agiv Tel Aviv Univerity 4/5/7 By Roy Ganor and Uri Juhaz Reference emantic with Application, H. Nielon and F. Nielon,

More information

A Practical Model for Minimizing Waiting Time in a Transit Network

A Practical Model for Minimizing Waiting Time in a Transit Network A Practical Model for Minimizing Waiting Time in a Tranit Network Leila Dianat, MASc, Department of Civil Engineering, Sharif Univerity of Technology, Tehran, Iran Youef Shafahi, Ph.D. Aociate Profeor,

More information

Trainable Context Model for Multiscale Segmentation

Trainable Context Model for Multiscale Segmentation Trainable Context Model for Multicale Segmentation Hui Cheng and Charle A. Bouman School of Electrical and Computer Engineering Purdue Univerity Wet Lafayette, IN 47907-1285 {hui, bouman}@ ecn.purdue.edu

More information

Laboratory Exercise 6

Laboratory Exercise 6 Laboratory Exercie 6 Adder, Subtractor, and Multiplier The purpoe of thi exercie i to examine arithmetic circuit that add, ubtract, and multiply number. Each type of circuit will be implemented in two

More information

A Boyer-Moore Approach for. Two-Dimensional Matching. Jorma Tarhio. University of California. Berkeley, CA Abstract

A Boyer-Moore Approach for. Two-Dimensional Matching. Jorma Tarhio. University of California. Berkeley, CA Abstract A Boyer-Moore Approach for Two-Dimenional Matching Jorma Tarhio Computer Science Diviion Univerity of California Berkeley, CA 94720 Abtract An imple ublinear algorithm i preented for two-dimenional tring

More information

Texture-Constrained Active Shape Models

Texture-Constrained Active Shape Models 107 Texture-Contrained Active Shape Model Shuicheng Yan, Ce Liu Stan Z. Li Hongjiang Zhang Heung-Yeung Shum Qianheng Cheng Microoft Reearch Aia, Beijing Sigma Center, Beijing 100080, China Dept. of Info.

More information

How to. write a paper. The basics writing a solid paper Different communities/different standards Common errors

How to. write a paper. The basics writing a solid paper Different communities/different standards Common errors How to write a paper The baic writing a olid paper Different communitie/different tandard Common error Reource Raibert eay My grammar point Article on a v. the Bug in writing Clarity Goal Conciene Calling

More information

Multi-Target Tracking In Clutter

Multi-Target Tracking In Clutter Multi-Target Tracking In Clutter John N. Sander-Reed, Mary Jo Duncan, W.B. Boucher, W. Michael Dimmler, Shawn O Keefe ABSTRACT A high frame rate (0 Hz), multi-target, video tracker ha been developed and

More information

Shortest Paths with Single-Point Visibility Constraint

Shortest Paths with Single-Point Visibility Constraint Shortet Path with Single-Point Viibility Contraint Ramtin Khoravi Mohammad Ghodi Department of Computer Engineering Sharif Univerity of Technology Abtract Thi paper tudie the problem of finding a hortet

More information

ES205 Analysis and Design of Engineering Systems: Lab 1: An Introductory Tutorial: Getting Started with SIMULINK

ES205 Analysis and Design of Engineering Systems: Lab 1: An Introductory Tutorial: Getting Started with SIMULINK ES05 Analyi and Deign of Engineering Sytem: Lab : An Introductory Tutorial: Getting Started with SIMULINK What i SIMULINK? SIMULINK i a oftware package for modeling, imulating, and analyzing dynamic ytem.

More information

Temporal Abstract Interpretation. To have a continuum of program analysis techniques ranging from model-checking to static analysis.

Temporal Abstract Interpretation. To have a continuum of program analysis techniques ranging from model-checking to static analysis. Temporal Abtract Interpretation Patrick COUSOT DI, École normale upérieure 45 rue d Ulm 75230 Pari cedex 05, France mailto:patrick.couot@en.fr http://www.di.en.fr/ couot and Radhia COUSOT LIX École polytechnique

More information

This article appeared in a journal published by Elsevier. The attached copy is furnished to the author for internal non-commercial research and

This article appeared in a journal published by Elsevier. The attached copy is furnished to the author for internal non-commercial research and Thi article appeared in a journal publihed by Elevier. The attached copy i furnihed to the author for internal non-commercial reearch and education ue, including for intruction at the author intitution

More information

Chapter S:II (continued)

Chapter S:II (continued) Chapter S:II (continued) II. Baic Search Algorithm Sytematic Search Graph Theory Baic State Space Search Depth-Firt Search Backtracking Breadth-Firt Search Uniform-Cot Search AND-OR Graph Baic Depth-Firt

More information

Aalborg Universitet. Published in: Proceedings of the Working Conference on Advanced Visual Interfaces

Aalborg Universitet. Published in: Proceedings of the Working Conference on Advanced Visual Interfaces Aalborg Univeritet Software-Baed Adjutment of Mobile Autotereocopic Graphic Uing Static Parallax Barrier Paprocki, Martin Marko; Krog, Kim Srirat; Kritofferen, Morten Bak; Krau, Martin Publihed in: Proceeding

More information

Compressed Sensing Image Processing Based on Stagewise Orthogonal Matching Pursuit

Compressed Sensing Image Processing Based on Stagewise Orthogonal Matching Pursuit Senor & randucer, Vol. 8, Iue 0, October 204, pp. 34-40 Senor & randucer 204 by IFSA Publihing, S. L. http://www.enorportal.com Compreed Sening Image Proceing Baed on Stagewie Orthogonal Matching Puruit

More information

AN ALGORITHM FOR RESTRICTED NORMAL FORM TO SOLVE DUAL TYPE NON-CANONICAL LINEAR FRACTIONAL PROGRAMMING PROBLEM

AN ALGORITHM FOR RESTRICTED NORMAL FORM TO SOLVE DUAL TYPE NON-CANONICAL LINEAR FRACTIONAL PROGRAMMING PROBLEM RAC Univerity Journal, Vol IV, No, 7, pp 87-9 AN ALGORITHM FOR RESTRICTED NORMAL FORM TO SOLVE DUAL TYPE NON-CANONICAL LINEAR FRACTIONAL PROGRAMMING PROLEM Mozzem Hoain Department of Mathematic Ghior Govt

More information

DAROS: Distributed User-Server Assignment And Replication For Online Social Networking Applications

DAROS: Distributed User-Server Assignment And Replication For Online Social Networking Applications DAROS: Ditributed Uer-Server Aignment And Replication For Online Social Networking Application Thuan Duong-Ba School of EECS Oregon State Univerity Corvalli, OR 97330, USA Email: duongba@eec.oregontate.edu

More information

Refining SIRAP with a Dedicated Resource Ceiling for Self-Blocking

Refining SIRAP with a Dedicated Resource Ceiling for Self-Blocking Refining SIRAP with a Dedicated Reource Ceiling for Self-Blocking Mori Behnam, Thoma Nolte Mälardalen Real-Time Reearch Centre P.O. Box 883, SE-721 23 Väterå, Sweden {mori.behnam,thoma.nolte}@mdh.e ABSTRACT

More information

Building a Compact On-line MRF Recognizer for Large Character Set using Structured Dictionary Representation and Vector Quantization Technique

Building a Compact On-line MRF Recognizer for Large Character Set using Structured Dictionary Representation and Vector Quantization Technique 202 International Conference on Frontier in Handwriting Recognition Building a Compact On-line MRF Recognizer for Large Character Set uing Structured Dictionary Repreentation and Vector Quantization Technique

More information

Shortest Paths Problem. CS 362, Lecture 20. Today s Outline. Negative Weights

Shortest Paths Problem. CS 362, Lecture 20. Today s Outline. Negative Weights Shortet Path Problem CS 6, Lecture Jared Saia Univerity of New Mexico Another intereting problem for graph i that of finding hortet path Aume we are given a weighted directed graph G = (V, E) with two

More information

A User-Attention Based Focus Detection Framework and Its Applications

A User-Attention Based Focus Detection Framework and Its Applications A Uer-Attention Baed Focu Detection Framework and It Application Chia-Chiang Ho, Wen-Huang Cheng, Ting-Jian Pan, Ja-Ling Wu Communication and Multimedia Laboratory, Department of Computer Science and Information

More information

Today s Outline. CS 561, Lecture 23. Negative Weights. Shortest Paths Problem. The presence of a negative cycle might mean that there is

Today s Outline. CS 561, Lecture 23. Negative Weights. Shortest Paths Problem. The presence of a negative cycle might mean that there is Today Outline CS 56, Lecture Jared Saia Univerity of New Mexico The path that can be trodden i not the enduring and unchanging Path. The name that can be named i not the enduring and unchanging Name. -

More information

Lecture Outline. Global flow analysis. Global Optimization. Global constant propagation. Liveness analysis. Local Optimization. Global Optimization

Lecture Outline. Global flow analysis. Global Optimization. Global constant propagation. Liveness analysis. Local Optimization. Global Optimization Lecture Outline Global flow analyi Global Optimization Global contant propagation Livene analyi Adapted from Lecture by Prof. Alex Aiken and George Necula (UCB) CS781(Praad) L27OP 1 CS781(Praad) L27OP

More information

SIMIT 7. Component Type Editor (CTE) User manual. Siemens Industrial

SIMIT 7. Component Type Editor (CTE) User manual. Siemens Industrial SIMIT 7 Component Type Editor (CTE) Uer manual Siemen Indutrial Edition January 2013 Siemen offer imulation oftware to plan, imulate and optimize plant and machine. The imulation- and optimizationreult

More information

/06/$ IEEE 364

/06/$ IEEE 364 006 IEEE International ympoium on ignal Proceing and Information Technology oie Variance Etimation In ignal Proceing David Makovoz IPAC, California Intitute of Technology, MC-0, Paadena, CA, 95 davidm@ipac.caltech.edu;

More information

IMPROVED JPEG DECOMPRESSION OF DOCUMENT IMAGES BASED ON IMAGE SEGMENTATION. Tak-Shing Wong, Charles A. Bouman, and Ilya Pollak

IMPROVED JPEG DECOMPRESSION OF DOCUMENT IMAGES BASED ON IMAGE SEGMENTATION. Tak-Shing Wong, Charles A. Bouman, and Ilya Pollak IMPROVED DECOMPRESSION OF DOCUMENT IMAGES BASED ON IMAGE SEGMENTATION Tak-Shing Wong, Charle A. Bouman, and Ilya Pollak School of Electrical and Computer Engineering Purdue Univerity ABSTRACT We propoe

More information

A Hybrid Deployable Dynamic Traffic Assignment Framework for Robust Online Route Guidance

A Hybrid Deployable Dynamic Traffic Assignment Framework for Robust Online Route Guidance A Hybrid Deployable Dynamic Traffic Aignment Framework for Robut Online Route Guidance Sriniva Peeta School of Civil Engineering, Purdue Univerity Chao Zhou Sabre, Inc. Sriniva Peeta School of Civil Engineering

More information

Interface Tracking in Eulerian and MMALE Calculations

Interface Tracking in Eulerian and MMALE Calculations Interface Tracking in Eulerian and MMALE Calculation Gabi Luttwak Rafael P.O.Box 2250, Haifa 31021,Irael Interface Tracking in Eulerian and MMALE Calculation 3D Volume of Fluid (VOF) baed recontruction

More information

Performance Evaluation of an Advanced Local Search Evolutionary Algorithm

Performance Evaluation of an Advanced Local Search Evolutionary Algorithm Anne Auger and Nikolau Hanen Performance Evaluation of an Advanced Local Search Evolutionary Algorithm Proceeding of the IEEE Congre on Evolutionary Computation, CEC 2005 c IEEE Performance Evaluation

More information

A Multi-objective Genetic Algorithm for Reliability Optimization Problem

A Multi-objective Genetic Algorithm for Reliability Optimization Problem International Journal of Performability Engineering, Vol. 5, No. 3, April 2009, pp. 227-234. RAMS Conultant Printed in India A Multi-objective Genetic Algorithm for Reliability Optimization Problem AMAR

More information

Localized Minimum Spanning Tree Based Multicast Routing with Energy-Efficient Guaranteed Delivery in Ad Hoc and Sensor Networks

Localized Minimum Spanning Tree Based Multicast Routing with Energy-Efficient Guaranteed Delivery in Ad Hoc and Sensor Networks Localized Minimum Spanning Tree Baed Multicat Routing with Energy-Efficient Guaranteed Delivery in Ad Hoc and Senor Network Hanne Frey Univerity of Paderborn D-3398 Paderborn hanne.frey@uni-paderborn.de

More information

A METHOD OF REAL-TIME NURBS INTERPOLATION WITH CONFINED CHORD ERROR FOR CNC SYSTEMS

A METHOD OF REAL-TIME NURBS INTERPOLATION WITH CONFINED CHORD ERROR FOR CNC SYSTEMS Vietnam Journal of Science and Technology 55 (5) (017) 650-657 DOI: 10.1565/55-518/55/5/906 A METHOD OF REAL-TIME NURBS INTERPOLATION WITH CONFINED CHORD ERROR FOR CNC SYSTEMS Nguyen Huu Quang *, Banh

More information

Implementation of a momentum-based distance metric for motion graphs. Student: Alessandro Di Domenico (st.no ), Supervisor: Nicolas Pronost

Implementation of a momentum-based distance metric for motion graphs. Student: Alessandro Di Domenico (st.no ), Supervisor: Nicolas Pronost Implementation of a momentum-baed ditance metric for motion graph Student: Aleandro Di Domenico (t.no 3775682), Supervior: Nicola Pronot April 3, 2014 Abtract Thi report preent the procedure and reult

More information

On successive packing approach to multidimensional (M-D) interleaving

On successive packing approach to multidimensional (M-D) interleaving On ucceive packing approach to multidimenional (M-D) interleaving Xi Min Zhang Yun Q. hi ankar Bau Abtract We propoe an interleaving cheme for multidimenional (M-D) interleaving. To achieved by uing a

More information

Design of a Stewart Platform for General Machining Using Magnetic Bearings

Design of a Stewart Platform for General Machining Using Magnetic Bearings eign of a Stewart Platform for eneral Machining Uing Magnetic earing Jeff Pieper epartment of Mechanical and Manufacturing Engineering Univerity of algary algary lberta anada N N4 pieper@ucalgary.ca Preented

More information

AUTOMATIC TEST CASE GENERATION USING UML MODELS

AUTOMATIC TEST CASE GENERATION USING UML MODELS Volume-2, Iue-6, June-2014 AUTOMATIC TEST CASE GENERATION USING UML MODELS 1 SAGARKUMAR P. JAIN, 2 KHUSHBOO S. LALWANI, 3 NIKITA K. MAHAJAN, 4 BHAGYASHREE J. GADEKAR 1,2,3,4 Department of Computer Engineering,

More information

New Structural Decomposition Techniques for Constraint Satisfaction Problems

New Structural Decomposition Techniques for Constraint Satisfaction Problems New Structural Decompoition Technique for Contraint Satifaction Problem Yaling Zheng and Berthe Y. Choueiry Contraint Sytem Laboratory Univerity of Nebraka-Lincoln Email: yzheng choueiry@ce.unl.edu Abtract.

More information

Security Analysis of the Efficient Chaos Pseudo-random Number Generator Applied to Video Encryption

Security Analysis of the Efficient Chaos Pseudo-random Number Generator Applied to Video Encryption Journal of Electronic Teting (2018) 34:709 715 http://doi.org/10.1007/10836-018-5767-0 Security Analyi of the Efficient Chao Peudo-random Number Generator Applied to Video Encryption Dragan Lambić 1 Alekandar

More information

A Linear Interpolation-Based Algorithm for Path Planning and Replanning on Girds *

A Linear Interpolation-Based Algorithm for Path Planning and Replanning on Girds * Advance in Linear Algebra & Matrix Theory, 2012, 2, 20-24 http://dx.doi.org/10.4236/alamt.2012.22003 Publihed Online June 2012 (http://www.scirp.org/journal/alamt) A Linear Interpolation-Baed Algorithm

More information

Karen L. Collins. Wesleyan University. Middletown, CT and. Mark Hovey MIT. Cambridge, MA Abstract

Karen L. Collins. Wesleyan University. Middletown, CT and. Mark Hovey MIT. Cambridge, MA Abstract Mot Graph are Edge-Cordial Karen L. Collin Dept. of Mathematic Weleyan Univerity Middletown, CT 6457 and Mark Hovey Dept. of Mathematic MIT Cambridge, MA 239 Abtract We extend the definition of edge-cordial

More information

Domain-Specific Modeling for Rapid System-Wide Energy Estimation of Reconfigurable Architectures

Domain-Specific Modeling for Rapid System-Wide Energy Estimation of Reconfigurable Architectures Domain-Specific Modeling for Rapid Sytem-Wide Energy Etimation of Reconfigurable Architecture Seonil Choi 1,Ju-wookJang 2, Sumit Mohanty 1, Viktor K. Praanna 1 1 Dept. of Electrical Engg. 2 Dept. of Electronic

More information

KS3 Maths Assessment Objectives

KS3 Maths Assessment Objectives KS3 Math Aement Objective Tranition Stage 9 Ratio & Proportion Probabilit y & Statitic Appreciate the infinite nature of the et of integer, real and rational number Can interpret fraction and percentage

More information

Analyzing Hydra Historical Statistics Part 2

Analyzing Hydra Historical Statistics Part 2 Analyzing Hydra Hitorical Statitic Part Fabio Maimo Ottaviani EPV Technologie White paper 5 hnode HSM Hitorical Record The hnode i the hierarchical data torage management node and ha to perform all the

More information

Image authentication and tamper detection using fragile watermarking in spatial domain

Image authentication and tamper detection using fragile watermarking in spatial domain International Journal of Advanced Reearch in Computer Engineering & Technology (IJARCET) Volume 6, Iue 7, July 2017, ISSN: 2278 1323 Image authentication and tamper detection uing fragile watermarking

More information

Key Terms - MinMin, MaxMin, Sufferage, Task Scheduling, Standard Deviation, Load Balancing.

Key Terms - MinMin, MaxMin, Sufferage, Task Scheduling, Standard Deviation, Load Balancing. Volume 3, Iue 11, November 2013 ISSN: 2277 128X International Journal of Advanced Reearch in Computer Science and Software Engineering Reearch Paper Available online at: www.ijarce.com Tak Aignment in

More information

A note on degenerate and spectrally degenerate graphs

A note on degenerate and spectrally degenerate graphs A note on degenerate and pectrally degenerate graph Noga Alon Abtract A graph G i called pectrally d-degenerate if the larget eigenvalue of each ubgraph of it with maximum degree D i at mot dd. We prove

More information

Towards an Efficient Optimal Trajectory Planner for Multiple Mobile Robots

Towards an Efficient Optimal Trajectory Planner for Multiple Mobile Robots Toward an Efficient Optimal Trajectory Planner for Multiple Mobile Robot Jaon Thoma, Alan Blair, Nick Barne Department of Computer Science and Software Engineering The Univerity of Melbourne, Victoria,

More information

Modeling and Analysis of Slow CW Decrease for IEEE WLAN

Modeling and Analysis of Slow CW Decrease for IEEE WLAN Modeling and Analyi of Slow CW Decreae for IEEE 82. WLAN Qiang Ni, Imad Aad 2, Chadi Barakat, and Thierry Turletti Planete Group 2 Planete Group INRIA Sophia Antipoli INRIA Rhône-Alpe Sophia Antipoli,

More information

Maneuverable Relays to Improve Energy Efficiency in Sensor Networks

Maneuverable Relays to Improve Energy Efficiency in Sensor Networks Maneuverable Relay to Improve Energy Efficiency in Senor Network Stephan Eidenbenz, Luka Kroc, Jame P. Smith CCS-5, MS M997; Lo Alamo National Laboratory; Lo Alamo, NM 87545. Email: {eidenben, kroc, jpmith}@lanl.gov

More information

3D MODELLING WITH LINEAR APPROACHES USING GEOMETRIC PRIMITIVES

3D MODELLING WITH LINEAR APPROACHES USING GEOMETRIC PRIMITIVES MAKARA, TEKNOLOGI, VOL. 9, NO., APRIL 5: 3-35 3D MODELLING WITH LINEAR APPROACHES USING GEOMETRIC PRIMITIVES Mochammad Zulianyah Informatic Engineering, Faculty of Engineering, ARS International Univerity,

More information

Integration of Digital Test Tools to the Internet-Based Environment MOSCITO

Integration of Digital Test Tools to the Internet-Based Environment MOSCITO Integration of Digital Tet Tool to the Internet-Baed Environment MOSCITO Abtract Current paper decribe a new environment MOSCITO for providing acce to tool over the internet. The environment i built according

More information

A TOPSIS based Method for Gene Selection for Cancer Classification

A TOPSIS based Method for Gene Selection for Cancer Classification Volume 67 No17, April 2013 A TOPSIS baed Method for Gene Selection for Cancer Claification IMAbd-El Fattah,WIKhedr, KMSallam, 1 Department of Statitic, 3 Department of Deciion upport, 2 Department of information

More information

A PROBABILISTIC NOTION OF CAMERA GEOMETRY: CALIBRATED VS. UNCALIBRATED

A PROBABILISTIC NOTION OF CAMERA GEOMETRY: CALIBRATED VS. UNCALIBRATED A PROBABILISTIC NOTION OF CAMERA GEOMETRY: CALIBRATED VS. UNCALIBRATED Jutin Domke and Yianni Aloimono Computational Viion Laboratory, Center for Automation Reearch Univerity of Maryland College Park,

More information

Handling Degenerate Cases in Exact Geodesic Computation on Triangle Meshes

Handling Degenerate Cases in Exact Geodesic Computation on Triangle Meshes The Viual Computer manucript. (will be inerted b the editor) Yong-Jin Liu Qian-Yi Zhou Shi-Min Hu Degenerate Cae in Eact Geodeic Computation on Triangle Mehe Abtract The computation of eact geodeic on

More information

CERIAS Tech Report EFFICIENT PARALLEL ALGORITHMS FOR PLANAR st-graphs. by Mikhail J. Atallah, Danny Z. Chen, and Ovidiu Daescu

CERIAS Tech Report EFFICIENT PARALLEL ALGORITHMS FOR PLANAR st-graphs. by Mikhail J. Atallah, Danny Z. Chen, and Ovidiu Daescu CERIAS Tech Report 2003-15 EFFICIENT PARALLEL ALGORITHMS FOR PLANAR t-graphs by Mikhail J. Atallah, Danny Z. Chen, and Ovidiu Daecu Center for Education and Reearch in Information Aurance and Security,

More information

Comparison of Methods for Horizon Line Detection in Sea Images

Comparison of Methods for Horizon Line Detection in Sea Images Comparion of Method for Horizon Line Detection in Sea Image Tzvika Libe Evgeny Gerhikov and Samuel Koolapov Department of Electrical Engineering Braude Academic College of Engineering Karmiel 2982 Irael

More information

The norm Package. November 15, Title Analysis of multivariate normal datasets with missing values

The norm Package. November 15, Title Analysis of multivariate normal datasets with missing values The norm Package November 15, 2003 Verion 1.0-9 Date 2002/05/06 Title Analyi of multivariate normal dataet with miing value Author Ported to R by Alvaro A. Novo . Original by Joeph

More information

See chapter 8 in the textbook. Dr Muhammad Al Salamah, Industrial Engineering, KFUPM

See chapter 8 in the textbook. Dr Muhammad Al Salamah, Industrial Engineering, KFUPM Goal programming Objective of the topic: Indentify indutrial baed ituation where two or more objective function are required. Write a multi objective function model dla a goal LP Ue weighting um and preemptive

More information