XV International PhD Workshop OWD 2013, October Machine Learning for the Efficient Control of a Multi-Wheeled Mobile Robot

Size: px
Start display at page:

Download "XV International PhD Workshop OWD 2013, October Machine Learning for the Efficient Control of a Multi-Wheeled Mobile Robot"

Transcription

1 XV Internatonal PhD Workshop OWD 203, 9 22 October 203 Machne Learnng for the Effcent Control of a Mult-Wheeled Moble Robot Uladzmr Dzomn, Brest State Techncal Unversty (prof. Vladmr Golovko, Brest State Techncal Unversty) Abstract Ths paper presents an applcaton of the multagent renforcement learnng approach for the effcent control of a moble robot. Ths approach s based on a mult-agent system appled to mult-wheel control. The robot s platform s decomposed nto drvng modules agents that are traned ndependently. The proposed approach ncorporates multple Q-learnng agents, whch permts them to effectvely control every wheel relatve to other wheels. The power reward polcy wth common error reward s adjusted to produce effcent control. The proposed approach s appled for the dstrbuted control of a mult-wheel platform, n order to provde energy consumpton optmzaton.. Introducton An effcent robot control s one of the mportant tasks for the applcaton of moble robot n producton. The mportant control tasks are power consumpton optmzaton and optmal trajectory plannng. Control subsystems should provde energy consumpton optmzaton n a robot control system. The power consumpton problem s solved by motor power control optmzaton [], [2] and by effcent moton plannng [3]. The robot control subsystem cannot nfluence on motor parameters, but must have polcy for the effcent control (optmal speed parameter, maxmum start drvng power, safe slow down dstance). The trajectory plannng usually mplements by plannng subsystem [4], [5]. Such subsystem bulds trajectory and dvdes t nto dfferent parts, whch are reproduced by crcles and straght lnes. The robot control subsystem should provde movement along the trajectory parts. The problem of effcent control s an mportant prerequste for the applcaton of the moble robot platform that was developed n Hochschule Ravensburg-Wengarten. The 3D mage of the Robot s llustrated n the fg. a. Ths platform s based on four nnovated vehcle steerng modules [6]. The steerng module s shown n the fg. 2b and conssts of two wheels powered by separate motors and behaves lke a dfferental drve. It s mounted to the platform by means of a bearng whch allows unlmted rotaton of the module wth respect to the platform. The platform can contan three or more modules. a) b) Fg.. a) Robot platform 3D model; b) The drvng module. The problem of mult-agent control s researched usually as problem of formaton composton, trajectory plannng, dstrbuted control and others. In ths paper we consder problem of crcular moton for sngle and mult-module case. One soluton of ths problem [7] [9] s to calculate knematcs of one wheeled robot for crcle drvng and after generalze t for mult-vehcle system. Ths approach has shown good modelng results. The dsadvantage of ths technque s small flexblty and hgh computatonal complexty. Alternatve approach s to use coordnaton archtecture wth one or more leaders [0] where vrtual coordnate frame follows the crcle trajectory. The expermental results usng a multrobot platform have shown the effectveness of ths approach. The lmtaton of such a technque s explct leader requrement. In ths paper we solve problem of optmal control for mult-module case n cooperatve crcular moton. The objectve s to acheve a crcular moton around a vrtual reference beacon wth optmal forward and angular speed. In ths paper we develop renforcement learnng technque producng effcent control rule, based on the relatve pose of the module wth respect to the beacon. In order to llustrate the man features of the problem, the sngle module case and the mult-module scenaro followed one by one are examned. The key contrbuton of ths paper are purposed renforcement learnng model for robot postonng 397

2 that allow module postonng around beacon even f such beacon s dynamcally change ts poston, and purposed renforcement learnng model for multmodule scenaro that allow to adjustment module speed to requred value wthn mult-module platform. Ths model requre only postons of modules relatve to the center of the mult-module platform By combnng postonng and speed control modules are able to produce cooperatve control law for effcent crcular moton. As a result the developed system can be easy scaled for any number of modules. 2. Robot Control The conventonal approach for the platform control s knematcs calculaton and nverse knematcs modelng [6]. If any modules on platform wll be added or removed, t wll requre recalculaton of knematcs equatons and reconfguraton of control subsystem. Knematcs calculatons can apply only for symmetrc turnng. For example we cannot drve usng Ackerman (car-lke) drvng scheme [], because t sn t enough for the movng along the dffcult trajectory parts. Fg.3. Robot knematcs for the smmetrc turnng. It should be noted that the prevously developed control model uses only one rotaton center that les on the lne, whch s perpendcularly to the robot center (Fg. 3). The pont G s the rotaton center, the pont S s the robot center and the SG lne s the robot turnng radus. It s an mportant restrcton for the ndustral control system, because the robot cannot drve n others ways. 3. Steerng Module Agent Let s decompose the robot s platform nto the ndependent drvng module agents. Agent stays n physcal 2d envronment wth reference beacon as t s shown n Fg. 4. Beacon poston s defned by coordnates (x b, y b). Rotaton radus ρ s the dstance from the center of module to the beacon. Fg.4. State of the module wth respect to reference beacon. The angle error s calculated by the followng equatons: ϕ = arctan 2( x x, y y) () center ϕ err center b = ϕ ϕ (2) robot Here φ center and φ robot are known from envronment. In ths paper envronment s represented by physcal 2d model smulated by Player/State. Envronment provdes the all necessary nformaton about the agent and rotaton pont relatve postons. The envronment nformaton states are llustrated n the table. Envronmental nformaton Robot Get Value X robot poston, x Coordnate, m 2 Y robot poston, y Coordnate, m 3 X of beacon center, x b Coordnate, m 4 Y of beacon center, y b Coordnate, m 5 Robot orentaton angle, ϕ robot b Float number, radans -π < ϕ robot π Tab.. Beacon orentaton Float number, angle relatve to robot, radans 6 ϕ center -π < ϕ center π 7 Radus sze, ρ Float number, m The navgaton subsystem of real steerng uses odometer sensors for navgaton purposes n the presented platform. The full set of actons avalable to the agent s presented n the table 2. The agent can change the angle error φ err around beacon, usng control of lnear ν and angular speed ω. Agent actons Tab.2. Robot actons Value Encrease force, v m/s 2 Reduce force, v -0.0 m/s 3 Encrease turnng left, ω rad/s 4 Encrease turnng rght, ω -0.0 rad/s 5 Do nothng, 0 m/s, 0 rad/s 398

3 4. Mult-agent system of drvng modules One soluton of formaton control s the vrtual structure approach [0]. The basc dea s to specfy a vrtual leader or a vrtual coordnate frame located at the vrtual center of the formaton as a reference for the whole group such that each module s desred states can be defned relatve to the vrtual leader or the vrtual coordnate frame. Once the desred dynamcs of the vrtual structure are defned, then the desred moton for each agent s derved. As a result, sngle module path plannng and trajectory generaton technques can be employed for the vrtual leader or the vrtual coordnate frame whle trajectory trackng strateges can be employed for each module. Let, N steerng module agent s wth vrtual leader forms a mult-agent system called platform. Fg. 5 shows an llustratve example of such a structure wth a formaton composed of four modules, where (x b, y b) represents the beacon and C represents a vrtual coordnate frame located at a vrtual center (x c, y c) wth an orentaton φ c relatve to beacon and rotaton radus ρ c. Fg.5. The steerng modules platform. Platform contans addtonal nformaton such as square of platform and requred module topology ncludng ts desred postons relatve to the centrod of platform. Vrtual leader seen by envronment analogously to sngle steerng module agent t has a state, and can perform acton. It receves the same nformaton from envronment defned n the table I, and acton set defned n the table 2. It should be noted, that modules not drectly controlled by vrtual leader. The modules reman ndependent enttes and adopt ther behavor to conform desred poston n platform. In the fg. 6, (x, y ) and (x opt, y opt) represent, respectvely, the -th module s actual and desred poston, and d err represent the desred devaton of the -th module relatve to desred poston, where d err = d d (3) t opt Fg.6. State of the platform wth reference the -th module Here d t dstance from vrtual center to current module poston and d opt requred dstance between vrtual center and -th module poston derved from platform topology. The vrtual center poston s derved from centrod of platform area. The vrtual leader agent knows the goal optmal forward speed of whole platform lmted n bounds ν opt [ν opt mn, ν opt max], where ν opt mn and ν opt max s respectvely mnmum and maxmum values of optmal speed. 5. Cooperatve crcular moton problem Let s examne the control problem of crcular moton. The man objectve s to buld the control strategy so that all the modules wthn platform acheve crcular moton around the beacon, wth prescrbed radus of rotaton and dstances between neghbors. Then the task s to create cooperatve control rule for any confguraton of N modules wthn platform n crcular moton around a fxed beacon, wth rotaton radus ρ c defned for center of platform. Fg.7 depcts such a behavor. If each module can track ts desred poston accurately, then the desred formaton shape can be preserved accurately. Further requrements are to take nto account module postonng before movement, adaptaton of angular and lnear speed durng crcular movement. The control strategy should be scalable for varous numbers of agents and ts confguratons. The problem of fndng mult-agent control law for crcular moton can be decomposed n the followng steps (a) module postonng and (b) forward speed adjustment to ft desred radus and poston. For both problems we use renforcement learnng, whch permts to acheve generalzaton ablty. For example, the new beacon poston can be dynamcally assgned to platform, and the same control low can be used for module postonng. 6. Module postonng The secton dscusses a renforcement learnng method producng effcent control law for module orentaton around beacon. 399

4 6. Renforcement Learnng Framework Renforcement learnng (RL) s used as one of the technques to learn optmal control for autonomous agents n unknown envronment [2]. The renforcement learnng framework s shown n the Fg. 8. The man dea s that agent execute acton a t n partcular state s t, goes to the next state s t+ and receves numercal reward r t+ as a feedback of recent acton. Agent should explore state space and for every state fnd actons, whch s more rewarded than other n some fnte horzon. Acton of robot a t A ω s a change of angular speed ω t for gven moment of tme t. The learnng system s gven a postve reward when the robot orentaton closer to the goal orentaton (φ err t 0) usng optmal speed ω opt and a penalty when the orentaton of the robot devates from the correct or selected acton does not optmal for the gven poston. The value of the reward s defned as: r t t t = R( ϕ, ω ) (7) Where R s reward functon whch s represented by decson tree depctng n the fg 8. err Fg. 7. Renforcement learnng framework. Let, Q(s, a) s a Q-functon reflects qualty of selectng specfed acton a n state s. For gven Q-functon, the optmal acton a n specfed state s s defned as follows: a * = arg max Q( s, a) (4) a A( s) The ntal values of Q-functon are unknown and equal to zero. The learnng goal s to approxmate optmal Q-functon, e.g. fndng true Q-values for each acton n every state usng receved sequences of rewards durng state transtons. For the moment of tme t, the change of Q-value can be calculated as follows: Q ) αδ t t t ( s, a = (5) Where value α (0, ] s learnng rate, and δ t Temporal Dfference (TD) error. The agent s learned n such a way that the TD error s decreased. Usng Q-learnng rule [2], the temporal dfference error calculated by: t t t + t t δ = r γ max Q( s, a) Q( s, a ) (6) t+ a A( s ) Where r t reward value obtaned for acton α t selected n s t, and γ dscount rate, A(s t+ ) set of actons avalable at s t RL-model for Module Postonng Usng defned above Q-learnng rule defne more precsely RL-model for module postonng ncludng state, acton and reward functon descrpton. It can be formulated as learnng to fnd such a behavor polcy that mnmzes φ err. Let, state of agent wll be par of values s t = [φ err t, ω t ]. Acton set A ω = {Ø, ω +, ω -} s represented by value of angular speed from table II. Fg. 8. Reward functon decton tree. Here φ stop the value of angle, where robot reduce speed to stop at the correct orentaton, ω opt [ ] rad/s optmal speed mnmzng module power consumpton. Angular speed wthn ths range s gven the hghest award wth exceptons ncases of acceleraton and deceleraton. 7. Cooperatve Movng In ths secton, we consder a mult-agent renforcement learnng model for cooperatve movng problem. The problem s to control module s ndvdual speed n order to acheve stable crcular moton of whole platform. Modules wth dfferent dstances to beacon should have a dfferent speed: for two modules and j, wth dstances to beacon ρ and ρ j respectvely, the speed v j wll more than v f the dstance to beacon ρ j more than ρ. Every module should have addtonal polcy to control ts forward speed wth respect to speed of other modules. 7. Mult-Agent Renforcement Learnng Framework The man prncples of a such technque are descrbed n [3] [4]. The basc dea of selected approach s to use nfluences between module and platform vrtual leader to determne sequences of correct actons n order to coordnate behavor 400

5 among them. The good nfluences should be rewarded and negatve should be punshed. The code desgn queston s how to determne such nfluences n terms of receved ndvdual reward. RL-framework used for such control problem s llustrated n the fg. 9: represented by physcal-lke 2d world wth four modules, vrtual leader, platform descrpton and beacon poston. The preparng part of collectve movement smulaton s learnng of robot postonng. Ths step s done once for ndvdual module before any cooperatve smulaton sessons. The Learned polcy s stored and coped for other modules. The topology of Q-functon traned durng 720 epochs s shown n the fg. 0. Fg. 9. Mult-Agent RL framework. The -th module at the state s t selects acton α t usng current polcy Q and goes to next state s t+ takng acton to envronment. Platform observes changes done by executed acton, calculates and assgns reward r t+ to module as a feedback reflectng successveness of specfed acton. The same Q-learnng rule (5) (6) can be used to update module control polcy. The man dfference between both rules s that n second case reward s assgned by a vrtual leader nstead of envronment: t t t+ Q ( s, a ) = α[ rp + t+ t t γ max Q ( s, a) Q ( s, a )] t+ a A( s ) (8) Instead of tryng to buld global Q-functon Q({s, s 2,, s n}, {a, a 2,,a n}) for n modules we decompose the problem and buld set of local Q- functons Q (s, a), Q 2(s, a),, Q N(s, a), where every polcy contans specfc control rule for each module. The combnaton of such ndvdual polces produces cooperatve control law. 7.2 RL-model for cooperatve movng Let, state of the module s par of s t = {v t, d err}, where ν t current value of lnear speed, and d err t dstance error calculated by (8). Acton set A ν = {Ø, ν +, ν - } s represented by ncreasng/decreasng of lnear speed from the table II and acton a t A ν s a change of forward speed ν t for gven moment of tme t. The vrtual agent receves error nformaton for each module and calculates dsplacement error. Ths error can be postve (module ahead of the platform) or negatve (the module behnd of the platform). The learnng process follows toward to mnmzaton of d err for every module. The maxmum reward s gven for case where d err 0, and a penalty gven when the poston of the module devates from the predefned. 8. Smulaton results For smulaton purposes the Player/State [5] modelng envronment was used. Envronment s Fg. 0. Result topology of Q-functon. When the correct orentaton of the robot s determned, angular speed s calculated to ft the specfed drvng radus usng lnear speed produced by cooperatve control law: ω = v (0) ρ Fg. shows the platform ntal state (left) and postonng auto-adjustment (rght) usng learned polcy. Fg.. Intal and fnal agents poston. Fg. 2 shows the expermental result of cooperatve movement after learnng postonng. It takes 000 epochs n average. The external parameters of smulaton are summarzed n the table 3. In the case of modelng ω opt s chosen from predefned bounds to show the applcablty of proposed approach. For real robot, bounds of optmal speed s derved from documentaton. 40

6 Fg. 2. Agents team drvng process. External parameters Parameter Value 2 α, learnng rate 0.4 γ, dscount factor 0.7 ω, optmal speed 0.8 rad/s 3 opt ϕ 4 stop, angle for slow down 0.6 radans Tab Conclusons and future works The expermental results shows that descrbed mult-agent renforcement learnng framework can solve the problem of effcent mult-wheel robot control. The proposed approach ncorporates multple Q-learnng agents, whch permts them to effectvely control every wheel relatve to the vrtual leader. The reward functons desgned n order to produce effcent control. A vrtual leader s used to coordnate module speeds. Meanwhle, ths role could be assgned to any module whch has access to global nformaton on platform level. The advantages of ths method are follows: Decomposton means that the nstead of tryng to buld global Q-functon we buld a set of local Q- functons. Adaptablty the platform wll adapt ts behavor for dynamcally assgned beacon and wll auto reconfgure movng trajectory. Scalablty and generalzaton the same learnng technque s used for every agent, for every beacon poston and every platform confguraton. Acknowledgement I express sncere grattude to my advsors Prof. Vladmr Golovko and Prof. Ralf Stetter, to my colleague Anton Kabysh for deas and dscussng. Bblography [] J. C. Andreas. Energy-Effcent Electrc Motors. Marcel Dekker, 2 edton, 992. [2] P. Bertold, A. de Almeda, H. Falkner. Energy Effcency Improvements n Electrc Motors and Drves. Berln: Sprnger, 997. [3] A. Barl, M. Ceresa, and C. Pars. Energy- Savng Moton Control for An Autonomous Moble Robot. Internatonal Symposum on Industral Electroncs, pp , 995. [4] D. J. Balkcom and M. T. Mason. Extremal Trajectores for Bounded Velocty Dfferental Drve Robots. ICRA, pp , [5] Yongguo Me, Yung-Hsang Lu, Y. Charle Hu, C.S. George Lee. Energy-Effcent Moton Plannng for Moble Robots. Robotcs and Automaton, [6] R. Stetter, P. Zemnak and A. Pachnsk. Development, Realzaton and Control of a Moble Robot. Research and Educaton n Robotcs-EUROBOT 200, Communcaton n Computer and Informaton Scence, Sprnger, vol. 56, pp , 20. [7] N. Ceccarell, M. D Marco, A. Garull, and A. Ganntrapan. Collectve crcular moton of mult-vehcle systems wth sensory lmtatons. Proceedngs of the 44th Conference on Decson and Control, Sevlle, Span, pp , December [8] N. Ceccarell, M. D Marco, A. Garull, and A. Ganntrapan. Collectve crcular moton of mult-vehcle systems. Automatca, 44(2): , [9] D. Benedettell, A. Garull, and A. Ganntrapan. Expermental valdaton of collectve crcular moton for nonholonomc mult-vehcle systems. Robotcs and Autonomous Systems, 58: , 200. [0] W. Ren, N. Sorensen. Dstrbuted coordnaton archtecture for mult-robot formaton control. Robotcs and Autonomous Systems, 56(4): , [] R. Segwart, Illah R. Nourbakhsh. Introducton to Autonomous Moble Robots. MIT Press, [2] Rchard S. Sutton, Andrew G. Barto.: Renforcement Learnng: An Introducton. MIT Press., 998. [3] A. Kabysh, V. Golovko. General model for organzng nteractons n mult-agent systems. Internatonal Journal of Computng, (3): , Sept [4] A. Kabysh, V. Golovko. Influence Learnng for Mult-Agent Systems Based on Renforcement Learnng. Internatonal Journal of Computng (): [5] Rchard Vaughan. Massvely Multple Robot Smulatons n Stage, Swarm Intellgence 2(2-4):89-208, Sprnger Authors: Mr. Vladmr Domn Brest State Techncal Unversty 267 Moskovskaja str Brest The Republc of Belarus tel. (+375) emal: spas.work@gmal.com 402

Parallelism for Nested Loops with Non-uniform and Flow Dependences

Parallelism for Nested Loops with Non-uniform and Flow Dependences Parallelsm for Nested Loops wth Non-unform and Flow Dependences Sam-Jn Jeong Dept. of Informaton & Communcaton Engneerng, Cheonan Unversty, 5, Anseo-dong, Cheonan, Chungnam, 330-80, Korea. seong@cheonan.ac.kr

More information

A mathematical programming approach to the analysis, design and scheduling of offshore oilfields

A mathematical programming approach to the analysis, design and scheduling of offshore oilfields 17 th European Symposum on Computer Aded Process Engneerng ESCAPE17 V. Plesu and P.S. Agach (Edtors) 2007 Elsever B.V. All rghts reserved. 1 A mathematcal programmng approach to the analyss, desgn and

More information

Learning the Kernel Parameters in Kernel Minimum Distance Classifier

Learning the Kernel Parameters in Kernel Minimum Distance Classifier Learnng the Kernel Parameters n Kernel Mnmum Dstance Classfer Daoqang Zhang 1,, Songcan Chen and Zh-Hua Zhou 1* 1 Natonal Laboratory for Novel Software Technology Nanjng Unversty, Nanjng 193, Chna Department

More information

NUMERICAL SOLVING OPTIMAL CONTROL PROBLEMS BY THE METHOD OF VARIATIONS

NUMERICAL SOLVING OPTIMAL CONTROL PROBLEMS BY THE METHOD OF VARIATIONS ARPN Journal of Engneerng and Appled Scences 006-017 Asan Research Publshng Network (ARPN). All rghts reserved. NUMERICAL SOLVING OPTIMAL CONTROL PROBLEMS BY THE METHOD OF VARIATIONS Igor Grgoryev, Svetlana

More information

User Authentication Based On Behavioral Mouse Dynamics Biometrics

User Authentication Based On Behavioral Mouse Dynamics Biometrics User Authentcaton Based On Behavoral Mouse Dynamcs Bometrcs Chee-Hyung Yoon Danel Donghyun Km Department of Computer Scence Department of Computer Scence Stanford Unversty Stanford Unversty Stanford, CA

More information

Smoothing Spline ANOVA for variable screening

Smoothing Spline ANOVA for variable screening Smoothng Splne ANOVA for varable screenng a useful tool for metamodels tranng and mult-objectve optmzaton L. Rcco, E. Rgon, A. Turco Outlne RSM Introducton Possble couplng Test case MOO MOO wth Game Theory

More information

Integration of Planning and Control in Robotic Formations

Integration of Planning and Control in Robotic Formations Integraton of Plannng and Control n Robotc Formatons V.T. Ngo, A.D. Nguyen, and Q.P. Ha ARC Centre of Excellence for Autonomous Systems, Faculty of Engneerng, Unversty of Technology, Sydney, PO Box 13

More information

Support Vector Machines

Support Vector Machines /9/207 MIST.6060 Busness Intellgence and Data Mnng What are Support Vector Machnes? Support Vector Machnes Support Vector Machnes (SVMs) are supervsed learnng technques that analyze data and recognze patterns.

More information

The Codesign Challenge

The Codesign Challenge ECE 4530 Codesgn Challenge Fall 2007 Hardware/Software Codesgn The Codesgn Challenge Objectves In the codesgn challenge, your task s to accelerate a gven software reference mplementaton as fast as possble.

More information

R s s f. m y s. SPH3UW Unit 7.3 Spherical Concave Mirrors Page 1 of 12. Notes

R s s f. m y s. SPH3UW Unit 7.3 Spherical Concave Mirrors Page 1 of 12. Notes SPH3UW Unt 7.3 Sphercal Concave Mrrors Page 1 of 1 Notes Physcs Tool box Concave Mrror If the reflectng surface takes place on the nner surface of the sphercal shape so that the centre of the mrror bulges

More information

SLAM Summer School 2006 Practical 2: SLAM using Monocular Vision

SLAM Summer School 2006 Practical 2: SLAM using Monocular Vision SLAM Summer School 2006 Practcal 2: SLAM usng Monocular Vson Javer Cvera, Unversty of Zaragoza Andrew J. Davson, Imperal College London J.M.M Montel, Unversty of Zaragoza. josemar@unzar.es, jcvera@unzar.es,

More information

Range images. Range image registration. Examples of sampling patterns. Range images and range surfaces

Range images. Range image registration. Examples of sampling patterns. Range images and range surfaces Range mages For many structured lght scanners, the range data forms a hghly regular pattern known as a range mage. he samplng pattern s determned by the specfc scanner. Range mage regstraton 1 Examples

More information

Positive Semi-definite Programming Localization in Wireless Sensor Networks

Positive Semi-definite Programming Localization in Wireless Sensor Networks Postve Sem-defnte Programmng Localzaton n Wreless Sensor etworks Shengdong Xe 1,, Jn Wang, Aqun Hu 1, Yunl Gu, Jang Xu, 1 School of Informaton Scence and Engneerng, Southeast Unversty, 10096, anjng Computer

More information

Load Balancing for Hex-Cell Interconnection Network

Load Balancing for Hex-Cell Interconnection Network Int. J. Communcatons, Network and System Scences,,, - Publshed Onlne Aprl n ScRes. http://www.scrp.org/journal/jcns http://dx.do.org/./jcns.. Load Balancng for Hex-Cell Interconnecton Network Saher Manaseer,

More information

A Binarization Algorithm specialized on Document Images and Photos

A Binarization Algorithm specialized on Document Images and Photos A Bnarzaton Algorthm specalzed on Document mages and Photos Ergna Kavalleratou Dept. of nformaton and Communcaton Systems Engneerng Unversty of the Aegean kavalleratou@aegean.gr Abstract n ths paper, a

More information

Skew Angle Estimation and Correction of Hand Written, Textual and Large areas of Non-Textual Document Images: A Novel Approach

Skew Angle Estimation and Correction of Hand Written, Textual and Large areas of Non-Textual Document Images: A Novel Approach Angle Estmaton and Correcton of Hand Wrtten, Textual and Large areas of Non-Textual Document Images: A Novel Approach D.R.Ramesh Babu Pyush M Kumat Mahesh D Dhannawat PES Insttute of Technology Research

More information

Overview. Basic Setup [9] Motivation and Tasks. Modularization 2008/2/20 IMPROVED COVERAGE CONTROL USING ONLY LOCAL INFORMATION

Overview. Basic Setup [9] Motivation and Tasks. Modularization 2008/2/20 IMPROVED COVERAGE CONTROL USING ONLY LOCAL INFORMATION Overvew 2 IMPROVED COVERAGE CONTROL USING ONLY LOCAL INFORMATION Introducton Mult- Smulator MASIM Theoretcal Work and Smulaton Results Concluson Jay Wagenpfel, Adran Trachte Motvaton and Tasks Basc Setup

More information

Outline. Discriminative classifiers for image recognition. Where in the World? A nearest neighbor recognition example 4/14/2011. CS 376 Lecture 22 1

Outline. Discriminative classifiers for image recognition. Where in the World? A nearest neighbor recognition example 4/14/2011. CS 376 Lecture 22 1 4/14/011 Outlne Dscrmnatve classfers for mage recognton Wednesday, Aprl 13 Krsten Grauman UT-Austn Last tme: wndow-based generc obect detecton basc ppelne face detecton wth boostng as case study Today:

More information

Mathematics 256 a course in differential equations for engineering students

Mathematics 256 a course in differential equations for engineering students Mathematcs 56 a course n dfferental equatons for engneerng students Chapter 5. More effcent methods of numercal soluton Euler s method s qute neffcent. Because the error s essentally proportonal to the

More information

Cluster Analysis of Electrical Behavior

Cluster Analysis of Electrical Behavior Journal of Computer and Communcatons, 205, 3, 88-93 Publshed Onlne May 205 n ScRes. http://www.scrp.org/ournal/cc http://dx.do.org/0.4236/cc.205.350 Cluster Analyss of Electrcal Behavor Ln Lu Ln Lu, School

More information

Improvement of Spatial Resolution Using BlockMatching Based Motion Estimation and Frame. Integration

Improvement of Spatial Resolution Using BlockMatching Based Motion Estimation and Frame. Integration Improvement of Spatal Resoluton Usng BlockMatchng Based Moton Estmaton and Frame Integraton Danya Suga and Takayuk Hamamoto Graduate School of Engneerng, Tokyo Unversty of Scence, 6-3-1, Nuku, Katsuska-ku,

More information

Reciprocal Collision Avoidance for Multiple Car-like Robots

Reciprocal Collision Avoidance for Multiple Car-like Robots Recprocal Collson Avodance for Multple Car-lke Robots Javer Alonso-Mora, Andreas Bretenmoser, Paul Beardsley and Roland Segwart Abstract In ths paper a method for dstrbuted recprocal collson avodance among

More information

Real-time Motion Capture System Using One Video Camera Based on Color and Edge Distribution

Real-time Motion Capture System Using One Video Camera Based on Color and Edge Distribution Real-tme Moton Capture System Usng One Vdeo Camera Based on Color and Edge Dstrbuton YOSHIAKI AKAZAWA, YOSHIHIRO OKADA, AND KOICHI NIIJIMA Graduate School of Informaton Scence and Electrcal Engneerng,

More information

X- Chart Using ANOM Approach

X- Chart Using ANOM Approach ISSN 1684-8403 Journal of Statstcs Volume 17, 010, pp. 3-3 Abstract X- Chart Usng ANOM Approach Gullapall Chakravarth 1 and Chaluvad Venkateswara Rao Control lmts for ndvdual measurements (X) chart are

More information

A high precision collaborative vision measurement of gear chamfering profile

A high precision collaborative vision measurement of gear chamfering profile Internatonal Conference on Advances n Mechancal Engneerng and Industral Informatcs (AMEII 05) A hgh precson collaboratve vson measurement of gear chamferng profle Conglng Zhou, a, Zengpu Xu, b, Chunmng

More information

Decision Support for the Dynamic Reconfiguration of Machine Layout and Part Routing in Cellular Manufacturing

Decision Support for the Dynamic Reconfiguration of Machine Layout and Part Routing in Cellular Manufacturing Decson Support for the Dynamc Reconfguraton of Machne Layout and Part Routng n Cellular Manufacturng Hao W. Ln and Tomohro Murata Abstract A mathematcal based approach s presented to evaluate the dynamc

More information

Simulation Based Analysis of FAST TCP using OMNET++

Simulation Based Analysis of FAST TCP using OMNET++ Smulaton Based Analyss of FAST TCP usng OMNET++ Umar ul Hassan 04030038@lums.edu.pk Md Term Report CS678 Topcs n Internet Research Sprng, 2006 Introducton Internet traffc s doublng roughly every 3 months

More information

Classifier Selection Based on Data Complexity Measures *

Classifier Selection Based on Data Complexity Measures * Classfer Selecton Based on Data Complexty Measures * Edth Hernández-Reyes, J.A. Carrasco-Ochoa, and J.Fco. Martínez-Trndad Natonal Insttute for Astrophyscs, Optcs and Electroncs, Lus Enrque Erro No.1 Sta.

More information

Content Based Image Retrieval Using 2-D Discrete Wavelet with Texture Feature with Different Classifiers

Content Based Image Retrieval Using 2-D Discrete Wavelet with Texture Feature with Different Classifiers IOSR Journal of Electroncs and Communcaton Engneerng (IOSR-JECE) e-issn: 78-834,p- ISSN: 78-8735.Volume 9, Issue, Ver. IV (Mar - Apr. 04), PP 0-07 Content Based Image Retreval Usng -D Dscrete Wavelet wth

More information

Optimal Design of Nonlinear Fuzzy Model by Means of Independent Fuzzy Scatter Partition

Optimal Design of Nonlinear Fuzzy Model by Means of Independent Fuzzy Scatter Partition Optmal Desgn of onlnear Fuzzy Model by Means of Independent Fuzzy Scatter Partton Keon-Jun Park, Hyung-Kl Kang and Yong-Kab Km *, Department of Informaton and Communcaton Engneerng, Wonkwang Unversty,

More information

Active Contours/Snakes

Active Contours/Snakes Actve Contours/Snakes Erkut Erdem Acknowledgement: The sldes are adapted from the sldes prepared by K. Grauman of Unversty of Texas at Austn Fttng: Edges vs. boundares Edges useful sgnal to ndcate occludng

More information

Problem Definitions and Evaluation Criteria for Computational Expensive Optimization

Problem Definitions and Evaluation Criteria for Computational Expensive Optimization Problem efntons and Evaluaton Crtera for Computatonal Expensve Optmzaton B. Lu 1, Q. Chen and Q. Zhang 3, J. J. Lang 4, P. N. Suganthan, B. Y. Qu 6 1 epartment of Computng, Glyndwr Unversty, UK Faclty

More information

The motion simulation of three-dof parallel manipulator based on VBAI and MATLAB Zhuo Zhen, Chaoying Liu* and Xueling Song

The motion simulation of three-dof parallel manipulator based on VBAI and MATLAB Zhuo Zhen, Chaoying Liu* and Xueling Song Internatonal Conference on Automaton, Mechancal Control and Computatonal Engneerng (AMCCE 25) he moton smulaton of three-dof parallel manpulator based on VBAI and MALAB Zhuo Zhen, Chaoyng Lu* and Xuelng

More information

Compiler Design. Spring Register Allocation. Sample Exercises and Solutions. Prof. Pedro C. Diniz

Compiler Design. Spring Register Allocation. Sample Exercises and Solutions. Prof. Pedro C. Diniz Compler Desgn Sprng 2014 Regster Allocaton Sample Exercses and Solutons Prof. Pedro C. Dnz USC / Informaton Scences Insttute 4676 Admralty Way, Sute 1001 Marna del Rey, Calforna 90292 pedro@s.edu Regster

More information

Investigations of Topology and Shape of Multi-material Optimum Design of Structures

Investigations of Topology and Shape of Multi-material Optimum Design of Structures Advanced Scence and Tecnology Letters Vol.141 (GST 2016), pp.241-245 ttp://dx.do.org/10.14257/astl.2016.141.52 Investgatons of Topology and Sape of Mult-materal Optmum Desgn of Structures Quoc Hoan Doan

More information

A Saturation Binary Neural Network for Crossbar Switching Problem

A Saturation Binary Neural Network for Crossbar Switching Problem A Saturaton Bnary Neural Network for Crossbar Swtchng Problem Cu Zhang 1, L-Qng Zhao 2, and Rong-Long Wang 2 1 Department of Autocontrol, Laonng Insttute of Scence and Technology, Benx, Chna bxlkyzhangcu@163.com

More information

Multi-stable Perception. Necker Cube

Multi-stable Perception. Necker Cube Mult-stable Percepton Necker Cube Spnnng dancer lluson, Nobuuk Kaahara Fttng and Algnment Computer Vson Szelsk 6.1 James Has Acknowledgment: Man sldes from Derek Hoem, Lana Lazebnk, and Grauman&Lebe 2008

More information

IP Camera Configuration Software Instruction Manual

IP Camera Configuration Software Instruction Manual IP Camera 9483 - Confguraton Software Instructon Manual VBD 612-4 (10.14) Dear Customer, Wth your purchase of ths IP Camera, you have chosen a qualty product manufactured by RADEMACHER. Thank you for the

More information

Sum of Linear and Fractional Multiobjective Programming Problem under Fuzzy Rules Constraints

Sum of Linear and Fractional Multiobjective Programming Problem under Fuzzy Rules Constraints Australan Journal of Basc and Appled Scences, 2(4): 1204-1208, 2008 ISSN 1991-8178 Sum of Lnear and Fractonal Multobjectve Programmng Problem under Fuzzy Rules Constrants 1 2 Sanjay Jan and Kalash Lachhwan

More information

Virtual Machine Migration based on Trust Measurement of Computer Node

Virtual Machine Migration based on Trust Measurement of Computer Node Appled Mechancs and Materals Onlne: 2014-04-04 ISSN: 1662-7482, Vols. 536-537, pp 678-682 do:10.4028/www.scentfc.net/amm.536-537.678 2014 Trans Tech Publcatons, Swtzerland Vrtual Machne Mgraton based on

More information

Development of a FPGA-based Motion Control IC for Robot Arm

Development of a FPGA-based Motion Control IC for Robot Arm Development of a FPGA-based oton Control IC for Robot Arm Yng-Sheh ung, ember IEEE Southern Tawan Unversty of Technology No. Nan-Ta Street, Yung-ang Tanan County, 7, TAIWAN kung@mal.stut.edu.tw Gua-Sheh

More information

Course Introduction. Algorithm 8/31/2017. COSC 320 Advanced Data Structures and Algorithms. COSC 320 Advanced Data Structures and Algorithms

Course Introduction. Algorithm 8/31/2017. COSC 320 Advanced Data Structures and Algorithms. COSC 320 Advanced Data Structures and Algorithms Course Introducton Course Topcs Exams, abs, Proects A quc loo at a few algorthms 1 Advanced Data Structures and Algorthms Descrpton: We are gong to dscuss algorthm complexty analyss, algorthm desgn technques

More information

Machine Learning: Algorithms and Applications

Machine Learning: Algorithms and Applications 14/05/1 Machne Learnng: Algorthms and Applcatons Florano Zn Free Unversty of Bozen-Bolzano Faculty of Computer Scence Academc Year 011-01 Lecture 10: 14 May 01 Unsupervsed Learnng cont Sldes courtesy of

More information

Support Vector Machines

Support Vector Machines Support Vector Machnes Decson surface s a hyperplane (lne n 2D) n feature space (smlar to the Perceptron) Arguably, the most mportant recent dscovery n machne learnng In a nutshell: map the data to a predetermned

More information

An Iterative Solution Approach to Process Plant Layout using Mixed Integer Optimisation

An Iterative Solution Approach to Process Plant Layout using Mixed Integer Optimisation 17 th European Symposum on Computer Aded Process Engneerng ESCAPE17 V. Plesu and P.S. Agach (Edtors) 2007 Elsever B.V. All rghts reserved. 1 An Iteratve Soluton Approach to Process Plant Layout usng Mxed

More information

Inverse kinematic Modeling of 3RRR Parallel Robot

Inverse kinematic Modeling of 3RRR Parallel Robot ème Congrès Franças de Mécanque Lyon, 4 au 8 Août 5 Inverse knematc Modelng of RRR Parallel Robot Ouafae HAMDOUN, Fatma Zahra BAGHLI, Larb EL BAKKALI Modelng and Smulaton of Mechancal Systems Laboratory,

More information

Wishing you all a Total Quality New Year!

Wishing you all a Total Quality New Year! Total Qualty Management and Sx Sgma Post Graduate Program 214-15 Sesson 4 Vnay Kumar Kalakband Assstant Professor Operatons & Systems Area 1 Wshng you all a Total Qualty New Year! Hope you acheve Sx sgma

More information

A New Approach For the Ranking of Fuzzy Sets With Different Heights

A New Approach For the Ranking of Fuzzy Sets With Different Heights New pproach For the ankng of Fuzzy Sets Wth Dfferent Heghts Pushpnder Sngh School of Mathematcs Computer pplcatons Thapar Unversty, Patala-7 00 Inda pushpndersnl@gmalcom STCT ankng of fuzzy sets plays

More information

The Greedy Method. Outline and Reading. Change Money Problem. Greedy Algorithms. Applications of the Greedy Strategy. The Greedy Method Technique

The Greedy Method. Outline and Reading. Change Money Problem. Greedy Algorithms. Applications of the Greedy Strategy. The Greedy Method Technique //00 :0 AM Outlne and Readng The Greedy Method The Greedy Method Technque (secton.) Fractonal Knapsack Problem (secton..) Task Schedulng (secton..) Mnmum Spannng Trees (secton.) Change Money Problem Greedy

More information

Solving two-person zero-sum game by Matlab

Solving two-person zero-sum game by Matlab Appled Mechancs and Materals Onlne: 2011-02-02 ISSN: 1662-7482, Vols. 50-51, pp 262-265 do:10.4028/www.scentfc.net/amm.50-51.262 2011 Trans Tech Publcatons, Swtzerland Solvng two-person zero-sum game by

More information

A Fast Visual Tracking Algorithm Based on Circle Pixels Matching

A Fast Visual Tracking Algorithm Based on Circle Pixels Matching A Fast Vsual Trackng Algorthm Based on Crcle Pxels Matchng Zhqang Hou hou_zhq@sohu.com Chongzhao Han czhan@mal.xjtu.edu.cn Ln Zheng Abstract: A fast vsual trackng algorthm based on crcle pxels matchng

More information

An Optimal Algorithm for Prufer Codes *

An Optimal Algorithm for Prufer Codes * J. Software Engneerng & Applcatons, 2009, 2: 111-115 do:10.4236/jsea.2009.22016 Publshed Onlne July 2009 (www.scrp.org/journal/jsea) An Optmal Algorthm for Prufer Codes * Xaodong Wang 1, 2, Le Wang 3,

More information

Related-Mode Attacks on CTR Encryption Mode

Related-Mode Attacks on CTR Encryption Mode Internatonal Journal of Network Securty, Vol.4, No.3, PP.282 287, May 2007 282 Related-Mode Attacks on CTR Encrypton Mode Dayn Wang, Dongda Ln, and Wenlng Wu (Correspondng author: Dayn Wang) Key Laboratory

More information

An Image Fusion Approach Based on Segmentation Region

An Image Fusion Approach Based on Segmentation Region Rong Wang, L-Qun Gao, Shu Yang, Yu-Hua Cha, and Yan-Chun Lu An Image Fuson Approach Based On Segmentaton Regon An Image Fuson Approach Based on Segmentaton Regon Rong Wang, L-Qun Gao, Shu Yang 3, Yu-Hua

More information

APPLICATION OF MULTIVARIATE LOSS FUNCTION FOR ASSESSMENT OF THE QUALITY OF TECHNOLOGICAL PROCESS MANAGEMENT

APPLICATION OF MULTIVARIATE LOSS FUNCTION FOR ASSESSMENT OF THE QUALITY OF TECHNOLOGICAL PROCESS MANAGEMENT 3. - 5. 5., Brno, Czech Republc, EU APPLICATION OF MULTIVARIATE LOSS FUNCTION FOR ASSESSMENT OF THE QUALITY OF TECHNOLOGICAL PROCESS MANAGEMENT Abstract Josef TOŠENOVSKÝ ) Lenka MONSPORTOVÁ ) Flp TOŠENOVSKÝ

More information

Lobachevsky State University of Nizhni Novgorod. Polyhedron. Quick Start Guide

Lobachevsky State University of Nizhni Novgorod. Polyhedron. Quick Start Guide Lobachevsky State Unversty of Nzhn Novgorod Polyhedron Quck Start Gude Nzhn Novgorod 2016 Contents Specfcaton of Polyhedron software... 3 Theoretcal background... 4 1. Interface of Polyhedron... 6 1.1.

More information

SURFACE PROFILE EVALUATION BY FRACTAL DIMENSION AND STATISTIC TOOLS USING MATLAB

SURFACE PROFILE EVALUATION BY FRACTAL DIMENSION AND STATISTIC TOOLS USING MATLAB SURFACE PROFILE EVALUATION BY FRACTAL DIMENSION AND STATISTIC TOOLS USING MATLAB V. Hotař, A. Hotař Techncal Unversty of Lberec, Department of Glass Producng Machnes and Robotcs, Department of Materal

More information

OPTIMAL CONFIGURATION FOR NODES IN MIXED CELLULAR AND MOBILE AD HOC NETWORK FOR INET

OPTIMAL CONFIGURATION FOR NODES IN MIXED CELLULAR AND MOBILE AD HOC NETWORK FOR INET OPTIMAL CONFIGURATION FOR NODE IN MIED CELLULAR AND MOBILE AD HOC NETWORK FOR INET Olusola Babalola D.E. Department of Electrcal and Computer Engneerng Morgan tate Unversty Dr. Rchard Dean Faculty Advsor

More information

Analysis on the Workspace of Six-degrees-of-freedom Industrial Robot Based on AutoCAD

Analysis on the Workspace of Six-degrees-of-freedom Industrial Robot Based on AutoCAD Analyss on the Workspace of Sx-degrees-of-freedom Industral Robot Based on AutoCAD Jn-quan L 1, Ru Zhang 1,a, Fang Cu 1, Q Guan 1 and Yang Zhang 1 1 School of Automaton, Bejng Unversty of Posts and Telecommuncatons,

More information

3. CR parameters and Multi-Objective Fitness Function

3. CR parameters and Multi-Objective Fitness Function 3 CR parameters and Mult-objectve Ftness Functon 41 3. CR parameters and Mult-Objectve Ftness Functon 3.1. Introducton Cogntve rados dynamcally confgure the wreless communcaton system, whch takes beneft

More information

CS246: Mining Massive Datasets Jure Leskovec, Stanford University

CS246: Mining Massive Datasets Jure Leskovec, Stanford University CS46: Mnng Massve Datasets Jure Leskovec, Stanford Unversty http://cs46.stanford.edu /19/013 Jure Leskovec, Stanford CS46: Mnng Massve Datasets, http://cs46.stanford.edu Perceptron: y = sgn( x Ho to fnd

More information

Term Weighting Classification System Using the Chi-square Statistic for the Classification Subtask at NTCIR-6 Patent Retrieval Task

Term Weighting Classification System Using the Chi-square Statistic for the Classification Subtask at NTCIR-6 Patent Retrieval Task Proceedngs of NTCIR-6 Workshop Meetng, May 15-18, 2007, Tokyo, Japan Term Weghtng Classfcaton System Usng the Ch-square Statstc for the Classfcaton Subtask at NTCIR-6 Patent Retreval Task Kotaro Hashmoto

More information

A Fast Content-Based Multimedia Retrieval Technique Using Compressed Data

A Fast Content-Based Multimedia Retrieval Technique Using Compressed Data A Fast Content-Based Multmeda Retreval Technque Usng Compressed Data Borko Furht and Pornvt Saksobhavvat NSF Multmeda Laboratory Florda Atlantc Unversty, Boca Raton, Florda 3343 ABSTRACT In ths paper,

More information

A MOVING MESH APPROACH FOR SIMULATION BUDGET ALLOCATION ON CONTINUOUS DOMAINS

A MOVING MESH APPROACH FOR SIMULATION BUDGET ALLOCATION ON CONTINUOUS DOMAINS Proceedngs of the Wnter Smulaton Conference M E Kuhl, N M Steger, F B Armstrong, and J A Jones, eds A MOVING MESH APPROACH FOR SIMULATION BUDGET ALLOCATION ON CONTINUOUS DOMAINS Mark W Brantley Chun-Hung

More information

CHAPTER 2 PROPOSED IMPROVED PARTICLE SWARM OPTIMIZATION

CHAPTER 2 PROPOSED IMPROVED PARTICLE SWARM OPTIMIZATION 24 CHAPTER 2 PROPOSED IMPROVED PARTICLE SWARM OPTIMIZATION The present chapter proposes an IPSO approach for multprocessor task schedulng problem wth two classfcatons, namely, statc ndependent tasks and

More information

VFH*: Local Obstacle Avoidance with Look-Ahead Verification

VFH*: Local Obstacle Avoidance with Look-Ahead Verification 2000 IEEE Internatonal Conference on Robotcs and Automaton, San Francsco, CA, Aprl 24-28, 2000, pp. 2505-25 VFH*: Local Obstacle Avodance wth Look-Ahead Verfcaton Iwan Ulrch and Johann Borensten The Unversty

More information

Pose, Posture, Formation and Contortion in Kinematic Systems

Pose, Posture, Formation and Contortion in Kinematic Systems Pose, Posture, Formaton and Contorton n Knematc Systems J. Rooney and T. K. Tanev Department of Desgn and Innovaton, Faculty of Technology, The Open Unversty, Unted Kngdom Abstract. The concepts of pose,

More information

CS 534: Computer Vision Model Fitting

CS 534: Computer Vision Model Fitting CS 534: Computer Vson Model Fttng Sprng 004 Ahmed Elgammal Dept of Computer Scence CS 534 Model Fttng - 1 Outlnes Model fttng s mportant Least-squares fttng Maxmum lkelhood estmaton MAP estmaton Robust

More information

Irregular Cellular Learning Automata and Its Application to Clustering in Sensor Networks

Irregular Cellular Learning Automata and Its Application to Clustering in Sensor Networks Irregular Cellular Learnng Automata and Its Applcaton to Clusterng n Sensor Networks M. Esnaashar 1, M. R. Meybod 1,2 1 Soft Computng Laboratory, Computer Engneerng and Informaton Technology Department

More information

3-Wheel Swerve Drive - The Trouble with Tribots

3-Wheel Swerve Drive - The Trouble with Tribots 3-Wheel Swerve Drve - The Trouble wth Trbots Clem McKown - FRC Team 1640 17-August-2014 Executve Summary FRC's 2013 change n robot permeter rules (to 112 nch maxmum overall permeter from the earler maxmum

More information

Anytime Predictive Navigation of an Autonomous Robot

Anytime Predictive Navigation of an Autonomous Robot Anytme Predctve Navgaton of an Autonomous Robot Shu Yun Chung Department of Mechancal Engneerng Natonal Tawan Unversty Tape, Tawan Emal shuyun@robot0.me.ntu.edu.tw Abstract To acheve fully autonomous moble

More information

MECHATRONIC MODEL FOR THE DYNAMIC ANALYSIS OF A ROBOTIC SYSTEM

MECHATRONIC MODEL FOR THE DYNAMIC ANALYSIS OF A ROBOTIC SYSTEM Bulletn of the Translvana Unversty of Braşov Seres I: Engneerng Scences Vol. 6 (55) No. 2-2013 MECHATRONIC MODEL FOR THE DYNAMIC ANALYSIS OF A ROBOTIC SYSTEM Monca ENESCU 1 Abstract: The present paper

More information

Comparison of Heuristics for Scheduling Independent Tasks on Heterogeneous Distributed Environments

Comparison of Heuristics for Scheduling Independent Tasks on Heterogeneous Distributed Environments Comparson of Heurstcs for Schedulng Independent Tasks on Heterogeneous Dstrbuted Envronments Hesam Izakan¹, Ath Abraham², Senor Member, IEEE, Václav Snášel³ ¹ Islamc Azad Unversty, Ramsar Branch, Ramsar,

More information

Determining the Optimal Bandwidth Based on Multi-criterion Fusion

Determining the Optimal Bandwidth Based on Multi-criterion Fusion Proceedngs of 01 4th Internatonal Conference on Machne Learnng and Computng IPCSIT vol. 5 (01) (01) IACSIT Press, Sngapore Determnng the Optmal Bandwdth Based on Mult-crteron Fuson Ha-L Lang 1+, Xan-Mn

More information

A Real-time Local Path Planning Method Based on SVM for UGV

A Real-time Local Path Planning Method Based on SVM for UGV A Real-tme Local Path Plannng Method Based on SVM for UGV Chengchen Zhuge School of Computer Scence and Engneerng, Nanjng Unversty of Scence and Technology, Nanjng, Chna Zhenmn Tang* School of Computer

More information

3D vector computer graphics

3D vector computer graphics 3D vector computer graphcs Paolo Varagnolo: freelance engneer Padova Aprl 2016 Prvate Practce ----------------------------------- 1. Introducton Vector 3D model representaton n computer graphcs requres

More information

Kinematics Modeling and Analysis of MOTOMAN-HP20 Robot

Kinematics Modeling and Analysis of MOTOMAN-HP20 Robot nd Workshop on Advanced Research and Technolog n Industr Applcatons (WARTIA ) Knematcs Modelng and Analss of MOTOMAN-HP Robot Jou Fe, Chen Huang School of Mechancal Engneerng, Dalan Jaotong Unverst, Dalan,

More information

Numerical model describing optimization of fibres winding process on open and closed frame

Numerical model describing optimization of fibres winding process on open and closed frame Journal of Physcs: Conference Seres PAPER OPEN ACCESS Numercal model descrbng optmzaton of fbres wndng process on open and closed frame To cte ths artcle: M Petr et al 06 J Phys: Conf Ser 738 0094 Vew

More information

VISUAL SELECTION OF SURFACE FEATURES DURING THEIR GEOMETRIC SIMULATION WITH THE HELP OF COMPUTER TECHNOLOGIES

VISUAL SELECTION OF SURFACE FEATURES DURING THEIR GEOMETRIC SIMULATION WITH THE HELP OF COMPUTER TECHNOLOGIES UbCC 2011, Volume 6, 5002981-x manuscrpts OPEN ACCES UbCC Journal ISSN 1992-8424 www.ubcc.org VISUAL SELECTION OF SURFACE FEATURES DURING THEIR GEOMETRIC SIMULATION WITH THE HELP OF COMPUTER TECHNOLOGIES

More information

The Research of Ellipse Parameter Fitting Algorithm of Ultrasonic Imaging Logging in the Casing Hole

The Research of Ellipse Parameter Fitting Algorithm of Ultrasonic Imaging Logging in the Casing Hole Appled Mathematcs, 04, 5, 37-3 Publshed Onlne May 04 n ScRes. http://www.scrp.org/journal/am http://dx.do.org/0.436/am.04.584 The Research of Ellpse Parameter Fttng Algorthm of Ultrasonc Imagng Loggng

More information

Learning physical Models of Robots

Learning physical Models of Robots Learnng physcal Models of Robots Jochen Mück Technsche Unverstät Darmstadt jochen.mueck@googlemal.com Abstract In robotcs good physcal models are needed to provde approprate moton control for dfferent

More information

Helsinki University Of Technology, Systems Analysis Laboratory Mat Independent research projects in applied mathematics (3 cr)

Helsinki University Of Technology, Systems Analysis Laboratory Mat Independent research projects in applied mathematics (3 cr) Helsnk Unversty Of Technology, Systems Analyss Laboratory Mat-2.08 Independent research projects n appled mathematcs (3 cr) "! #$&% Antt Laukkanen 506 R ajlaukka@cc.hut.f 2 Introducton...3 2 Multattrbute

More information

Evaluation of an Enhanced Scheme for High-level Nested Network Mobility

Evaluation of an Enhanced Scheme for High-level Nested Network Mobility IJCSNS Internatonal Journal of Computer Scence and Network Securty, VOL.15 No.10, October 2015 1 Evaluaton of an Enhanced Scheme for Hgh-level Nested Network Moblty Mohammed Babker Al Mohammed, Asha Hassan.

More information

Sensory Redundant Parallel Mobile Mechanism

Sensory Redundant Parallel Mobile Mechanism Sensory Redundant Parallel Moble Mechansm Shraga Shoval and Moshe Shoham* Department of Industral Engneerng & Management, Academc College of Judea and Samara, Arel., *Faculty of Mechancal Engneerng, Technon,

More information

ON THE DESIGN OF LARGE SCALE REDUNDANT PARALLEL MANIPULATOR. Wu huapeng, Heikki handroos and Juha kilkki

ON THE DESIGN OF LARGE SCALE REDUNDANT PARALLEL MANIPULATOR. Wu huapeng, Heikki handroos and Juha kilkki ON THE DESIGN OF LARGE SCALE REDUNDANT PARALLEL MANIPULATOR Wu huapeng, Hekk handroos and Juha klkk Machne Automaton Lab, Lappeenranta Unversty of Technology LPR-5385 Fnland huapeng@lut.f, handroos@lut.f,

More information

INVERSE DYNAMICS ANALYSIS AND SIMULATION OF A CLASS OF UNDER- CONSTRAINED CABLE-DRIVEN PARALLEL SYSTEM

INVERSE DYNAMICS ANALYSIS AND SIMULATION OF A CLASS OF UNDER- CONSTRAINED CABLE-DRIVEN PARALLEL SYSTEM U.P.B. Sc. Bull., Seres D, Vol. 78, Iss., 6 ISSN 454-58 INVERSE DYNAMICS ANALYSIS AND SIMULATION OF A CLASS OF UNDER- CONSTRAINED CABLE-DRIVEN PARALLEL SYSTEM We LI, Zhgang ZHAO, Guangtan SHI, Jnsong LI

More information

Tsinghua University at TAC 2009: Summarizing Multi-documents by Information Distance

Tsinghua University at TAC 2009: Summarizing Multi-documents by Information Distance Tsnghua Unversty at TAC 2009: Summarzng Mult-documents by Informaton Dstance Chong Long, Mnle Huang, Xaoyan Zhu State Key Laboratory of Intellgent Technology and Systems, Tsnghua Natonal Laboratory for

More information

Robotic manipulation of multiple objects as a POMDP

Robotic manipulation of multiple objects as a POMDP Robotc manpulaton of multple objects as a POMDP Jon Pajarnen, Jon.Pajarnen@aalto.f Vlle Kyrk, Vlle.Kyrk@aalto.f Department of Electrcal Engneerng and Automaton, Aalto Unversty, Fnland arxv:1402.0649v2

More information

Performance Evaluation of Information Retrieval Systems

Performance Evaluation of Information Retrieval Systems Why System Evaluaton? Performance Evaluaton of Informaton Retreval Systems Many sldes n ths secton are adapted from Prof. Joydeep Ghosh (UT ECE) who n turn adapted them from Prof. Dk Lee (Unv. of Scence

More information

Edge Detection in Noisy Images Using the Support Vector Machines

Edge Detection in Noisy Images Using the Support Vector Machines Edge Detecton n Nosy Images Usng the Support Vector Machnes Hlaro Gómez-Moreno, Saturnno Maldonado-Bascón, Francsco López-Ferreras Sgnal Theory and Communcatons Department. Unversty of Alcalá Crta. Madrd-Barcelona

More information

Optimum Synthesis of Mechanisms For Path Generation Using a New Curvature Based Deflection Based Objective Function

Optimum Synthesis of Mechanisms For Path Generation Using a New Curvature Based Deflection Based Objective Function Proceedngs of the 6th WSEAS Internatonal Conference on Smulaton, Modellng and Optmzaton, Lsbon, Portugal, September -4, 6 67 Optmum Synthess of Mechansms For Path Generaton Usng a ew Curvature Based Deflecton

More information

Review of approximation techniques

Review of approximation techniques CHAPTER 2 Revew of appromaton technques 2. Introducton Optmzaton problems n engneerng desgn are characterzed by the followng assocated features: the objectve functon and constrants are mplct functons evaluated

More information

Fitting: Deformable contours April 26 th, 2018

Fitting: Deformable contours April 26 th, 2018 4/6/08 Fttng: Deformable contours Aprl 6 th, 08 Yong Jae Lee UC Davs Recap so far: Groupng and Fttng Goal: move from array of pxel values (or flter outputs) to a collecton of regons, objects, and shapes.

More information

Repeater Insertion for Two-Terminal Nets in Three-Dimensional Integrated Circuits

Repeater Insertion for Two-Terminal Nets in Three-Dimensional Integrated Circuits Repeater Inserton for Two-Termnal Nets n Three-Dmensonal Integrated Crcuts Hu Xu, Vasls F. Pavlds, and Govann De Mchel LSI - EPFL, CH-5, Swtzerland, {hu.xu,vasleos.pavlds,govann.demchel}@epfl.ch Abstract.

More information

Autonomous multi-vehicle formations for cooperative airfield snow shoveling

Autonomous multi-vehicle formations for cooperative airfield snow shoveling 1 Autonomous mult-vehcle formatons for cooperatve arfeld snow shovelng Martn Hess Martn Saska Klaus Schllng Unversty of Wuerzburg, Computer Scence VII: Robotcs and Telematcs, Germany Abstract In ths paper

More information

Optimizing Document Scoring for Query Retrieval

Optimizing Document Scoring for Query Retrieval Optmzng Document Scorng for Query Retreval Brent Ellwen baellwe@cs.stanford.edu Abstract The goal of ths project was to automate the process of tunng a document query engne. Specfcally, I used machne learnng

More information

Outline. Self-Organizing Maps (SOM) US Hebbian Learning, Cntd. The learning rule is Hebbian like:

Outline. Self-Organizing Maps (SOM) US Hebbian Learning, Cntd. The learning rule is Hebbian like: Self-Organzng Maps (SOM) Turgay İBRİKÇİ, PhD. Outlne Introducton Structures of SOM SOM Archtecture Neghborhoods SOM Algorthm Examples Summary 1 2 Unsupervsed Hebban Learnng US Hebban Learnng, Cntd 3 A

More information

Adaptive Energy and Location Aware Routing in Wireless Sensor Network

Adaptive Energy and Location Aware Routing in Wireless Sensor Network Adaptve Energy and Locaton Aware Routng n Wreless Sensor Network Hong Fu 1,1, Xaomng Wang 1, Yngshu L 1 Department of Computer Scence, Shaanx Normal Unversty, X an, Chna, 71006 fuhong433@gmal.com {wangxmsnnu@hotmal.cn}

More information

An Algorithm for Weighted Positive Influence Dominating Set Based on Learning Automata

An Algorithm for Weighted Positive Influence Dominating Set Based on Learning Automata 4 th Internatonal Conference on Knowledge-Based Engneerng and Innovaton (KBEI-2017) Dec. 22 th, 2017 (Iran Unversty of Scence and Technology) Tehran, Iran An Algorthm for Weghted Postve Influence Domnatng

More information

DESIGN OF A HAPTIC DEVICE FOR EXCAVATOR EQUIPPED WITH CRUSHER

DESIGN OF A HAPTIC DEVICE FOR EXCAVATOR EQUIPPED WITH CRUSHER DESIGN OF A HAPTIC DEVICE FOR EXCAVATOR EQUIPPED WITH CRUSHER Kyeong Won Oh, Dongnam Km Korea Unversty, Graduate School 5Ga-1, Anam-Dong, Sungbuk-Gu, Seoul, Korea {locosk, smleast}@korea.ac.kr Jong-Hyup

More information