Adaptive Flocking Control for Dynamic Target Tracking in Mobile Sensor Networks

Size: px
Start display at page:

Download "Adaptive Flocking Control for Dynamic Target Tracking in Mobile Sensor Networks"

Transcription

1 The 29 IEEE/RSJ Internatonal Conference on Intellgent Robots and Systems October -5, 29 St. Lous, USA Adaptve Flockng Control for Dynamc Target Trackng n Moble Sensor Networks Hung Manh La and Wehua Sheng Abstract Target trackng s an mportant task n sensor networks, especally n moble sensor networks. Flockng control s used to control a moble sensor network to track a target. However, there are some exstng problems n ths control method, such as network fragmentaton, loss of formaton and poor trackng performance. In order to handle these problems we propose a novel approach to flockng control of a moble sensor network to track a movng target wthn changng envronments. In our approach, each agent can cooperatvely learn the network s parameters to decde the sze of network n a decentralzed fashon so that the connectvty, formaton and trackng performance can be mproved when avodng obstacles. In addton, to demonstrate the beneft of our approach a comparson between ths approach and the exstng method s gven. Computer smulatons are performed to demonstrate the effectveness of the proposed approach. Keywords: Flockng, target trackng, moble sensor network. A. Motvaton I. INTRODUCTION Moble sensor networks [] have advantages over statonary sensor networks such as adaptaton to envronmental changes and reconfgurablty for better performance. A man ssue for multple moble sensors to track a movng target s that these sensors have to move together wthout collson among them durng trackng. Ths requres us to apply cooperatve control methods, and one of these methods s flockng control [2], [3], [4], [5]. However, these exstng works have some lmtatons when the envronment changes, for example when the moble sensor network has to pass through a narrow space among obstacles. These lmtatons nclude:. Connectvty s lost because of the fragmentaton phenomenon. 2. Formaton of the network s totally changed. 3. Low speed or gettng stuck causes poor trackng performance. Therefore to desgn an adaptve flockng control algorthm to deal wth these problems s a challengng task. In ths paper, we present a novel approach to flockng control of a moble sensor network to track a movng target n changng envronments. In ths approach, each agent cooperatvely learns the network s parameters n a decentralzed fashon Ths project s supported by the Vetnamese Government, MOET (Mnstry of Educaton and Tranng) program and the DoD ARO DURIP grant CS-RIP. Hung Manh La and Wehua Sheng are wth the school of Electrcal and Computer Engneerng, Oklahoma State Unversty, Stllwater, OK 7478, USA (hung.la@okstate.edu, wehua.sheng@okstate.edu). so that the connectvty, formaton and trackng performance can be mproved when avodng obstacles. The reason of mantanng the connectvty and smlar formaton s that when the network shrnks ts sze to deal wth changng envronments the neghborhood of each agent can be mantaned. Ths allows the network to keep the same topology, whch reduces the complexty of control durng the trackng process. Computer smulatons are conducted to prove our theoretcal results. B. Related work Flockng control has receved consderable attenton due to ts wde applcatons, such as space exploraton and survellance. In [2], the theoretcal framework for desgn and analyss of dstrbuted flockng algorthms was proposed. These algorthms solved the flockng n free space and n the presence of obstacles. The statc and dynamc vrtual leaders were used as a navgatonal feedback for all moble sensors. Ths establshed a background for flockng control desgn for a group of sensors. Adaptve flockng control, an extenson of flockng control, has also ganed attenton from researchers n recent years. Yang et al. [6] proposed an adaptve flockng control algorthm to avod collson among robots themselves and between robots and obstacles. However, ther algorthm dd not consder the problem of formaton, connectvty and trackng performance n complex envronments. In addton, ther algorthm only consdered a statc target or a rendezvous pont, whch leads all agents to get there. Lee and Chong [7] ntroduced a moton plannng framework for a large number of autonomous robots that enables the robots to confgure themselves adaptvely nto an area of arbtrary geometry. Ther proposed method allows the robots to converge to the unform dstrbuton by formng an equlateral trangle wth ther two neghbors. However, the problem of target trackng was not addressed n ther work. An extenson of ther work was developed n [8] by the same authors to allow the swarm of robots to go to predetermned rendezvous ponts. Ther approach was based on a decentralzed approach that enables a swarm of robots navgate autonomously n complex envronments populated by obstacles. The problem of splttng/mergng moble robots n the network accordng to the envronments s addressed n ther paper. Namely, when the swarm of robots detects obstacles, each robot splts from the network and determnes ts drecton toward the statc goal based on the wdth of space among obstacles. However, n realty t s dffcult for each robot to sense the whole envronment and compute the wdth of space among obstacles. Also, n ther work /9/$ IEEE 4843

2 the problem of controllng the sze of the network was not consdered, and the connectvty and formaton were not guaranteed n complex envronments. In summary, most of exstng work focused on the coordnaton, formaton and splttng/mergng problems n both fxed and swtchng topologes. The problem of how to control the sze of the network n a decentralzed and adaptve fashon n complex envronments whle mantanng connectvty, formaton and trackng performance s stll an open problem. The rest of ths paper s organzed as follows. In the next secton we present the background of flockng control. Secton III descrbes our adaptve flockng control algorthm to track a movng target whle avodng obstacles. Secton IV provdes the smulaton results. Fnally, secton V concludes ths paper. II. FLOCKING CONTROL BACKGROUND In ths secton we wll present the graph prelmnary and the flockng control background. We consder n sensors movng n an m (e.g.,m = 2,3) dmensonal Eucldean space. The dynamc equaton of each sensor s descrbed as follows: q = p () ṗ = u, =,2,...,n. To descrbe the topology of flocks or swarms we consder a dynamc graph G consstng of a vertex set ϑ =,2...,n} and an edge set E (, j) :, j ϑ, j }. In ths topology each vertex denotes one member of flocks, and each edge denotes the communcaton lnk between two members. Let q, p R m be the poston and velocty of node, respectvely. We know that durng the movement of sensors, the relatve dstance between them may change, hence the neghbors of each sensor also change. Therefore, we can defne a set of neghborhood of sensor as follows: N = j ϑ : q j q r, ϑ =,2,...,n}, j }, (2) here, r s an actve range (radus of neghborhood crcle n the case of two dmensons, m = 2, or radus of neghborhood sphere n the case of three dmensons, m = 3), and. s the Eucldean dstance. The geometry of flocks s modeled by an -lattce [2] that meets the followng condton: q j q = d, j N, (3) here d s a postve constant ndcatng the dstance between sensor and ts neghbor j. To construct a collectve potental that s dfferentable at sngular confguraton (q = q j ), the set of algebrac constrans s rewrtten n term of σ - norm as follows: q j q σ = d, j N, (4) here the constrant d = d σ wth d = r/k c, where k c s the scalng factor. The σ - norm [2],. σ, of a vector s a map R m = R + defned as z σ = /ε[ +ε z 2 ] wth ε >. Unlke the Eucldean norm z, whch s not dfferentable at z =, the σ - norm z σ, s dfferentable every where. Ths property allows to construct a smooth collectve potental functon for agents. The flockng control law n [2] controls all sensors to form an -lattce confguraton. Ths algorthm conssts of three components as follows: u = f + f β + f γ. (5) The frst component of (5) f, whch conssts of a gradent-based component and a consensus component (more detals about these components see [9], [], []), s used to regulate the potentals (mpulsve or attractve forces) and the velocty among sensors. f = c φ ( q j q σ )n j +c 2 a j (q)(p j p ), (6) j N j N where each term n (6) s computed as follows [2]: The acton functon φ (z) that vanshes for all z r wth r = r σ s defned as follows: φ (z) = ρ h (z/r )φ(z d ) (7) wth the uneven sgmodal functon φ(z) defned as φ(z) =.5[(a+b)σ (z+c)+(a b)], here σ (z) = z/ +z 2, and parameters < a b, c = a b / 4ab to guarantee φ() =. The bump functon ρ h (z) wth h (,), z [,h) ρ h (z) =.5[+cos(π( z h h ))], z [h,) (8), otherwse. Vector along the lne connectng q to q j s defned as n j = (q j q )/ +ε q j q 2. (9) The adjacency matrx a j (q) s defned as ρh ( q a j (q) = j q σ /r ), f j, f j =. () The second component of (5) f β s used to control the sensors to avod obstacles, f β = c β φ β ( ˆq,k q σ ) ˆn,k + c β 2 b,k (q)( ˆp,k p ), k N β k N β () where the set of β neghbors (vrtual neghbors, [2]) s } N β = j ϑ β : ˆq,k q r,ϑ β =,2,...,K}, (2) here K s the number of obstacles, r s an obstacle detectng range, and ˆq,k, ˆp,k are the poston and velocty of sensor projected on the obstacle k, respectvely (more detals please see [2]). Smlar to vector n j establshed n (9), vector ˆn,k s defned as ˆn,k = ( ˆq,k q )/ +ε ˆq,k q 2. (3) The heterogeneous adjacent matrx b,k (q) s defned as b,k (q) = ρ h ( ˆq,k q σ /d β ), (4) 4844

3 where d β = r σ. The repulsve acton functon of β neghbors s defned as φ β (z) = ρ h (z/d β )(σ (z d β ) ). (5) The thrd component of (5) f γ s a dstrbuted navgatonal feedback. f γ = c γ (q q γ ) c γ 2 (p p γ ), (6) where the γ - agent (q γ, p γ ) s the vrtual leader that leads the flock to follow ts trajectory, and t s defned as follows qγ = p γ (7) ṗ γ = f γ (q γ, p γ ). The constants of three components used n (5) are chosen as c < cγ < cβ, and cν 2 = 2 c ν. Here cν η are postve constants for η =,2 and ν =,β,γ. III. ADAPTIVE FLOCKING CONTROL FOR TRACKING A MOVING TARGET In ths paper, we consder the γ agent as a movng target. Hence, based on Olfat-Saber s flockng control [2] we desgn a control law wth a movng target as u = c φ ( q j q σ )n j + c 2 a j (q)(p j p ) j N j N +c β k N β φ β ( ˆq,k q σ ) ˆn,k + c β 2 b,k (q)( ˆp,k p ) k N β (q q mt ) c mt 2 (p p mt ), (8) here (q mt, p mt ) s the poston and velocty of the movng target, respectvely, and c mt are postve constants. In ths control law, we assume that each agent has ablty to sense the poston and velocty of the movng target. The problem here s how to cooperatvely control the sze of the network n an adaptve and decentralzed fashon n order to mantan the network s connectvty, smlar formaton and trackng performance n the presence of obstacles. One example of such flockng control s llustrated n Fgure. Fg.. Illustraton of the adaptve flockng control A. Adaptve flockng control To control the sze of the network, we need to control the set of algebrac constrants n Equaton (4), whch means that f we want the sze of the network to be smaller to pass the narrow space then d should be smaller. Ths rases the queston of how small the sze of network should be reduced and how to control the sze n a decentralzed and dynamc fashon. To control the constrant d one possble method s based on the knowledge of obstacle obtaned by any sensor n the network, whch wll broadcast a new d to all other sensors. However, t s dffcult for a sngle sensor to learn the sze of the obstacles due to ts lmted sensng range. To overcome ths problem we propose a method based on the repulsve force, β k N φ β ( ˆq,k q σ ), whch s generated by the β-agent (vrtual agent) projected on the obstacles. If any sensor n the network gets ths repulsve force t wll shrnk ts own d. If ths repulsve force s bg (sensor s close to obstacle(s)) d wll be further reduced. Then, n order to mantan the neghborhood (topology) the actve range of each sensor s re-desgned. To create the agreement on the relatve dstance and actve range among sensors n a decentralzed way, a consensus or a local average update law s proposed. Furthermore, to mantan the connectvty each sensor s desgned wth an adaptve weght of attractve force from the target and an adaptve weght of nteracton force from ts neghbors so that the network reduces or recovers the sze gradually. That s, f an sensor has weak connecton to the network t should have a bg weght of attracton force to the target and a small weght of nteracton force from ts neghbors. Frstly, we control the set of algebrac constrants as q j q σ = d, j N, (9) and let each agent have ts own d, whch s desgned as d, f d β k N φ β ( ˆq,k q σ ) = = c a β φ β ( ˆq,k q σ ) +, f k N β φ β ( ˆq,k q σ ), k N (2) here c a s the postve constant. From Equaton (2) we see that f the repulsve force generated from the obstacles β k N φ β ( ˆq,k q σ ) = or N β = / (empty set) then the agent wll keep ts orgnal d. When the agent senses the obstacles t reduces ts own d, and the value of d depends on the repulsve force that the agent gets from obstacles. In order to control the sze of network each sensor needs ts own r that relates to d as follows: r = k c d σ wth d σ = d (εd or d = +)2 ε. Explctly, r s computed as n Equaton (2). r, f β k N φ β ( ˆq,k q σ ) = r = ε [ k 2 (εd +)2 c ε + ], (2) f β k N φ β ( ˆq,k q σ ).

4 Smlar to computng r, r s computed as r, f β k N φ β ( ˆq,k q σ ) = r = ε [(εr + ) 2 ], f β k N φ β ( ˆq,k q σ ). (22) It should be ponted out that the actve range r s dfferent from the physcal communcaton (sensng) range. Namely, the actve range s the range that each agent decdes ts neghbors to talk wth, but the physcal communcaton range s the range defned by the RF module. Ths mples that even a robot can communcate wth many other robots n the network, t wll only talk (nteract) wth robots n ts actve range. That s why we want to control the actve range of each robot n order to reduce the communcaton and mantan the smlar formaton when the network shrnks. To acheve agreement on d, r and r among sensors n the connected network we use the followng update law based on local average for d, r and r : d = N + N + j= d j r = N + N + j= r j r = N + N + j= r j, (23) here N s the number of neghbors of agent. In addton, to better mantan the network connectvty each agent should have an adaptve weght of attractve force from the target and nteracton force from ts neghbors as dscussed before. Frstly, n the control protocol (8), the frst two terms are used to control the formaton (velocty matchng, collson avodance among robots). The thrd and fourth terms are used to allow robots to avod obstacles, and the last term s used for target trackng. If the last term s absent the control wll lead to network fragmentaton [2]. The coeffcents of the nteracton forces (c, c 2 ), (cβ, cβ 2 ) and attractve force (c mt ) whch delver desred swarmlke behavour are used to adjust the weght of nteracton forces and attractve force. The bgger (c mt ) the faster convergence to the target. However f (c mt ) s too bg the center of mass (CoM) as defned n Equaton (24) q = n n = q p = n n = p (24) oscllates around the target, and the formaton of network s not guaranteed. In addton, n order to guarantee that no agent ht obstacles, the par (c β, cβ 2 ) s selected to be bgger than the other two pars, (c, c 2 ) and (cmt ). Fnally we have the relatonshp among these pars as: (c,2 < cmt,2 < c β,2 ). From the above analyss we see that these adaptve weghts allow the network to reduce and recover the sze gradually. They also allow the network to mantan the connectvty durng the obstacle avodance. We let each sensor have ts own weght of the nteracton forces as n Equaton (25) and attractve force as n Equaton (26). In the -lattce confguraton f the sensor has less than 3 neghbors t s consdered as havng a weak connecton to the network. Ths means that ths sensor s on the border of network, or far from the target hence t should have bgger weght of attractve force from ts target and smaller weght of nteracton forces from ts neghbors to get closer to the target. Ths desgn also has the beneft of makng the whole network track the target faster. From ths analyss c,2 are desgned as follows: and cmt,2 of each agent c () = c, f N 3 c, f N < 3, (25) here c < c, c 2 () = 2 c (), and =,2,...,n. c mt () = c mt, f N 3 c mt, f N < 3, (26) here c mt > c mt () = 2 c mt (), and =,2,...,n. Now, the neghborhood of sensor (N ), the new adjacency matrx a j (q) and the new acton functon φ (z) are redefned as follows: N = j ϑ : q j q r, ϑ =,2,...,n}, j } ; (27) a j (q) = ρh ( q j q σ /r ), f j (28), f j = ; φ ( q j q σ )=ρ h ( q j q σ /r )φ( q j q σ d ). (29) Fnally, the adaptve flockng control law for dynamc target trackng s u = c () j N +c 2 () j N +c β k N β φ ( q j q σ )n j a j(q)(p j p ) φ β ( ˆq,k q σ ) ˆn,k + c β 2 b,k (q)( ˆp,k p ) k N β ()(q q mt ) c mt 2 ()(p p mt ). (3) B. Stablty Analyss By applyng the control protocol (3), the CoM (defned n Equaton (24)) of postons and veloctes of all moble sensors n the network wll exponentally converge to the target n both free space and obstacle space. In addton, the formaton (collson free and velocty matchng among moble sensors) wll mantan n the process of the target trackng. Let us consder adaptve flockng control n free space and obstacle space, respectvely. Case (Free space): In free space, β k N φ β ( ˆq,k q σ ) =, hence we can rewrte the control protocol (3) by gnorng constants c ν η (for η =,2 and ν =,β) as follows: u = q ψ ( q j q σ )+ a j (q)(p j p ) j N j N (q q mt ) c mt 2 (p p mt ) (3) 4846

5 where ψ (z) = z d φ (s)ds s the parwse attractve/repulsve potental functon. From (3), we can compute the average of control law u as follows: n u = = n =u n n = ( j N q ψ ( q j q σ ) + a j (q)(p j p )) j N (q q mt ) c mt 2 (p p mt ). (32) Obvously, we see that the par (ψ,a(q)) are symmetrc. Hence we can rewrte (32) as: u = (q q mt) c mt 2 (p p mt). (33) Equaton (33) mples that q = p ṗ = (q q mt) c mt 2 (p p mt). (34) The soluton of (34) ndcates that the CoM of postons and veloctes exponentally converge to those of the target. The formaton (collson-free and velocty matchng among moble sensors) s mantaned n the free space trackng because the gradent-based term and the consensus term are consdered n ths stuaton (more detals please see [2]). Case 2 (Obstacle space): Snce d s desgned to be reduced when each agent senses the obstacles. Therefore, when the sensor network have to pass through the narrow space between two obstacles ts sze wll be shrunk gradually, and when the network already passed ths narrow space t grows back to the orgnal sze gradually. Ths reduces the mpact of the obstacle on the network hence the speed of sensors can be mantaned or the CoM keeps trackng the target. Also, the connectvty and smlar formaton can be mantaned n ths scenaro. IV. SIMULATION RESULTS In ths secton we wll test our adaptve flockng control algorthm (3) and compare t wth the exstng flockng algorthm (8) n terms of the network connectvty, formaton and trackng performance. The parameters used n ths smulaton are specfed as follows: - Parameters of flockng: number of sensors = 5 (randomly dstrbuted n the box of x sze); a = b = 5; d = 7; the scalng factor k c =.2; the actve range r = k c d = 8.4; ε =. for the σ-norm; h =.2 for the bump functon (φ (z)); h =.9 for the bump functon (φ β (z)). - Parameters of target movement: The target moves n the lne trajectory: q mt = [+3t, t] T wth t 3.5, and p mt = (q mt (t) q mt (t ))/ t wth step sze t =.2. To analyze the connectvty of the network we defne a connectvty matrx c j (t) as follows:, f j N c j (t) = (t), j, f j N (t), j (35) and c =. Because the rank of Laplacan of a connected graph [2] c j (t) of order n s at most (n ) or rank(c j (t)) (n ), the relatve connectvty of a network at tme t s defned as C(t) = n rank(c j(t)). (36) If C(t) < the network s broken, and f C(t) = the network s connected. Based on ths metrc we can evaluate the network connectvty n our adaptve flockng control algorthm (3). Fgures 2 represents the results of movng target (red/dark lne) trackng n the lne trajectory usng the exstng flockng control algorthm (5). Fgures 3 represents the results of movng target trackng n the lne trajectory usng the adaptve flockng control algorthm (3). Fgure 4 shows the results of velocty matchng among sensors (a, a ), connectvty (b, b ) and error postons between the CoM (black/darker lne) and the target (trackng performance) (c, c ) of both flockng control algorthms (3) and (5), respectvely. To compare these algorthms we use the same ntal state (poston and velocty) of moble sensors. By comparng these fgures we see that by applyng the adaptve flockng control algorthm (3) the connectvty, smlar formaton and trackng performance are mantaned when the network passes through the narrow space between two obstacles (two red/dark crcles) whle the exstng flockng control algorthm (5) could not handle these problems. In Fgures 3 when the network enters the small gap between two obstacles ts sze s shrunk gradually n order to pass ths space, then the network sze grows back gradually when t passed. Therefore the connectvty and smlar formaton are mantaned. V. CONCLUSION Ths paper studed the approach to flockng control of a moble sensor network to track and observe a movng target n changng envronments. We desgned an adaptve flockng control algorthm that can cooperatvely learn the network s parameters n a decentralzed fashon to change the sze of the network n order to mantan connectvty, formaton and trackng performance when passng through obstacles. In addton, to see the beneft of the adaptve flockng algorthm we compared t wth the normal flockng control algorthm, and we found that the connectvty, smlar formaton and trackng performance n the adaptve flockng control algorthm are better than those n the exstng flockng control algorthm. The computer smulaton verfed our theoretcal results. REFERENCES [] S. Kamath, E. Mesner, and V. Isler. Trangulaton based mult target trackng wth moble sensor networks. IEEE Internatonal Conference on Robotcs and Automaton, pages , 27. [2] R. Olfat-Saber. Flockng for mult-agent dynamc systems: Algorthms and theory. IEEE Transactons on Automatc Control, 5(3):4 42, 26. [3] R. Olfat-Saber. Dstrbuted trackng for moble sensor networks wth nformaton drven moblty. Proceedngs of the 27 Amercan Control Conference,, pages ,

6 Fg. 2. Snapshots of the moble sensor network (a) when the moble sensors form a network, (b) when the moble sensors avod obstacles, (c) when the moble sensors get stuck n the narrow space between two obstacles. These results are obtaned by usng algorthm (5) Fg. 3. Snapshots of the moble sensor network (a) when the moble sensors form a network, (b) when the moble sensors avod obstacles, (c) when the moble sensors successfully passed through the narrow space between two obstacles, (d) when the moble sensors recover the orgnal sze. (a, b, c, d ) are closer look of (a, b, c, d), respectvely. These results are obtaned by usng algorthm (3) Fg. 4. Velocty matchng among sensors, connectvty, and error of postons between CoM and the movng target n (a, b, c) usng algorthm (3), (a, b, c ) usng algorthm (5), respectvely. [4] H. G. Tanner, A. Jadbaba, and G. J. Pappas. Stable flockng of moble agents, part : fxed topology. Proceedngs of the 42nd IEEE Conference on Decson and Control, pages 2 25, 23. [5] H. G. Tanner, A. Jadbaba, and G. J. Pappas. Stable flockng of moble agents, part : dynamc topology. Proceedngs of the 42nd IEEE Conference on Decson and Control, pages 26 22, 23. [6] Y. Yang, N. Xong, N. Y. Chong, and X. Defago. A decentralzed and adaptve flockng algorthm for autonomous moble robots. The 3rd Internatonal Conference on Grd and Pervasve Computng Workshops, IEEE computer socety, pages , 28. [7] G. Lee and N. Y. Chong. Adaptve self-confgurable robot swarms based on local nteractons. Proceedngs of the 27 IEEE/RSJ Internatonal Conference on Intellgent Robots and Systems, pages , 27. [8] G. Lee and N. Y. Chong. Adaptve flockng of robot swarms: Algorthms and propertes. IEICE transactons on Communcatons, (9): , 28. [9] P. Ogren, E. Forell, and N. E. Leonard. Cooperatve control of moble sensor networks: Adaptve gradent clmbng n a dstrbuted envronment. IEEE Transactons on Automatc Control, 49(8):292 32, 26. [] R. Olfat-Saber and R. M. Murray. Consensus problems n networks of agents wth swtchng topology and tme delays. IEEE Transactons on Automatc Control, 49(9):52 533, 24. [] R. Olfat-Saber, J. Alex Fax, and R. M. Murray. Consensus and cooperatve n networked mult-agent systems. Proceedngs of the IEEE, 95():25 233,

Flocking Control of a Mobile Sensor Network to Track and Observe a Moving Target

Flocking Control of a Mobile Sensor Network to Track and Observe a Moving Target 29 IEEE Internatonal Conference on Robotcs and Automaton Kobe Internatonal Conference Center Kobe, Japan, May 12-17, 29 Flockng Control of a Moble Sensor Network to Track and Observe a Movng Target Hung

More information

Adaptive Consensus Control of Multi-Agent Systems with Large Uncertainty and Time Delays *

Adaptive Consensus Control of Multi-Agent Systems with Large Uncertainty and Time Delays * Journal of Robotcs, etworkng and Artfcal Lfe, Vol., o. (September 04), 5-9 Adaptve Consensus Control of Mult-Agent Systems wth Large Uncertanty and me Delays * L Lu School of Mechancal Engneerng Unversty

More information

DO NOT DO HOMEWORK UNTIL IT IS ASSIGNED. THE ASSIGNMENTS MAY CHANGE UNTIL ANNOUNCED.

DO NOT DO HOMEWORK UNTIL IT IS ASSIGNED. THE ASSIGNMENTS MAY CHANGE UNTIL ANNOUNCED. EE 539 Homeworks Sprng 08 Updated: Tuesday, Aprl 7, 08 DO NOT DO HOMEWORK UNTIL IT IS ASSIGNED. THE ASSIGNMENTS MAY CHANGE UNTIL ANNOUNCED. For full credt, show all work. Some problems requre hand calculatons.

More information

Outline. Communication. Bellman Ford Algorithm. Bellman Ford Example. Bellman Ford Shortest Path [1]

Outline. Communication. Bellman Ford Algorithm. Bellman Ford Example. Bellman Ford Shortest Path [1] DYNAMIC SHORTEST PATH SEARCH AND SYNCHRONIZED TASK SWITCHING Jay Wagenpfel, Adran Trachte 2 Outlne Shortest Communcaton Path Searchng Bellmann Ford algorthm Algorthm for dynamc case Modfcatons to our algorthm

More information

Research Article Design of Connectivity Preserving Flocking Using Control Lyapunov Function

Research Article Design of Connectivity Preserving Flocking Using Control Lyapunov Function Journal of Robotcs Volume 1, Artcle ID 8571, 1 pages http://dx.do.org/1.1155/1/8571 Research Artcle Desgn of Connectvty Preservng Flockng Usng Control Lyapunov Functon Bayu Erfanto, 1, Ryanto T. Bambang,

More information

Distributed Exponential Formation Control of Multiple Wheeled Mobile Robots

Distributed Exponential Formation Control of Multiple Wheeled Mobile Robots Proceedngs of the Internatonal Conference of Control, Dynamc Systems, and Robotcs Ottawa, Ontaro, Canada, May 15-16 214 Paper No. 46 Dstrbuted Exponental Formaton Control of Multple Wheeled Moble Robots

More information

Pattern Generation with Multiple Robots

Pattern Generation with Multiple Robots Pattern Generaton wth Multple Robots Mong-yng A. Hseh and Vjay Kumar GRASP Laboratory Unversty of Pennsylvana Phladelpha, PA 1914 Emal: {mya, kumar}@grasp.cs.upenn.edu Abstract We develop decentralzed

More information

Kernel Methods and SVMs Extension

Kernel Methods and SVMs Extension Kernel Methods and SVMs Extenson The purpose of ths document s to revew materal covered n Machne Learnng 1 Supervsed Learnng regardng support vector machnes (SVMs). Ths document also provdes a general

More information

This document is downloaded from DR-NTU, Nanyang Technological University Library, Singapore.

This document is downloaded from DR-NTU, Nanyang Technological University Library, Singapore. Ths document s downloaded from DR-NTU Nanyang Technologcal Unversty Lbrary Sngapore. Ttle A New Navgaton Functon Based Decentralzed Control of Mult-Vehcle Systems n Unknown Envronments Author(s) Wang Yuanzhe;

More information

Module 3 LOSSY IMAGE COMPRESSION SYSTEMS. Version 2 ECE IIT, Kharagpur

Module 3 LOSSY IMAGE COMPRESSION SYSTEMS. Version 2 ECE IIT, Kharagpur Module 3 LOSSY IMAGE COMPRESSION SYSTEMS Verson ECE IIT, Kharagpur Lesson 6 Theory of Quantzaton Verson ECE IIT, Kharagpur Instructonal Objectves At the end of ths lesson, the students should be able to:

More information

Lecture 10: May 6, 2013

Lecture 10: May 6, 2013 TTIC/CMSC 31150 Mathematcal Toolkt Sprng 013 Madhur Tulsan Lecture 10: May 6, 013 Scrbe: Wenje Luo In today s lecture, we manly talked about random walk on graphs and ntroduce the concept of graph expander,

More information

ECE559VV Project Report

ECE559VV Project Report ECE559VV Project Report (Supplementary Notes Loc Xuan Bu I. MAX SUM-RATE SCHEDULING: THE UPLINK CASE We have seen (n the presentaton that, for downlnk (broadcast channels, the strategy maxmzng the sum-rate

More information

Power law and dimension of the maximum value for belief distribution with the max Deng entropy

Power law and dimension of the maximum value for belief distribution with the max Deng entropy Power law and dmenson of the maxmum value for belef dstrbuton wth the max Deng entropy Bngy Kang a, a College of Informaton Engneerng, Northwest A&F Unversty, Yanglng, Shaanx, 712100, Chna. Abstract Deng

More information

Physics 5153 Classical Mechanics. D Alembert s Principle and The Lagrangian-1

Physics 5153 Classical Mechanics. D Alembert s Principle and The Lagrangian-1 P. Guterrez Physcs 5153 Classcal Mechancs D Alembert s Prncple and The Lagrangan 1 Introducton The prncple of vrtual work provdes a method of solvng problems of statc equlbrum wthout havng to consder the

More information

The equation of motion of a dynamical system is given by a set of differential equations. That is (1)

The equation of motion of a dynamical system is given by a set of differential equations. That is (1) Dynamcal Systems Many engneerng and natural systems are dynamcal systems. For example a pendulum s a dynamcal system. State l The state of the dynamcal system specfes t condtons. For a pendulum n the absence

More information

CHALMERS, GÖTEBORGS UNIVERSITET. SOLUTIONS to RE-EXAM for ARTIFICIAL NEURAL NETWORKS. COURSE CODES: FFR 135, FIM 720 GU, PhD

CHALMERS, GÖTEBORGS UNIVERSITET. SOLUTIONS to RE-EXAM for ARTIFICIAL NEURAL NETWORKS. COURSE CODES: FFR 135, FIM 720 GU, PhD CHALMERS, GÖTEBORGS UNIVERSITET SOLUTIONS to RE-EXAM for ARTIFICIAL NEURAL NETWORKS COURSE CODES: FFR 35, FIM 72 GU, PhD Tme: Place: Teachers: Allowed materal: Not allowed: January 2, 28, at 8 3 2 3 SB

More information

Distributed Cooperative Control System Algorithms Simulations and Enhancements

Distributed Cooperative Control System Algorithms Simulations and Enhancements Po Wu and P.J. Antsakls, Dstrbuted Cooperatve Control System Algorthms: Smulatons and Enhancements, ISIS Techncal Report, Unversty of Notre Dame, ISIS-2009-001, Aprl 2009. (http://www.nd.edu/~ss/tech.html)

More information

Report on Image warping

Report on Image warping Report on Image warpng Xuan Ne, Dec. 20, 2004 Ths document summarzed the algorthms of our mage warpng soluton for further study, and there s a detaled descrpton about the mplementaton of these algorthms.

More information

DISTRIBUTED SENSOR FUSION USING DYNAMIC CONSENSUS. California Institute of Technology

DISTRIBUTED SENSOR FUSION USING DYNAMIC CONSENSUS. California Institute of Technology DISTRIBUTED SENSOR FUSION USING DYNAMIC CONSENSUS Demetr P. Spanos Rchard M. Murray Calforna Insttute of Technology Abstract: Ths work s an extenson to a companon paper descrbng consensustrackng for networked

More information

Markov Chain Monte Carlo Lecture 6

Markov Chain Monte Carlo Lecture 6 where (x 1,..., x N ) X N, N s called the populaton sze, f(x) f (x) for at least one {1, 2,..., N}, and those dfferent from f(x) are called the tral dstrbutons n terms of mportance samplng. Dfferent ways

More information

Supporting Information

Supporting Information Supportng Informaton The neural network f n Eq. 1 s gven by: f x l = ReLU W atom x l + b atom, 2 where ReLU s the element-wse rectfed lnear unt, 21.e., ReLUx = max0, x, W atom R d d s the weght matrx to

More information

Econ107 Applied Econometrics Topic 3: Classical Model (Studenmund, Chapter 4)

Econ107 Applied Econometrics Topic 3: Classical Model (Studenmund, Chapter 4) I. Classcal Assumptons Econ7 Appled Econometrcs Topc 3: Classcal Model (Studenmund, Chapter 4) We have defned OLS and studed some algebrac propertes of OLS. In ths topc we wll study statstcal propertes

More information

Inductance Calculation for Conductors of Arbitrary Shape

Inductance Calculation for Conductors of Arbitrary Shape CRYO/02/028 Aprl 5, 2002 Inductance Calculaton for Conductors of Arbtrary Shape L. Bottura Dstrbuton: Internal Summary In ths note we descrbe a method for the numercal calculaton of nductances among conductors

More information

An Emergent Wall Following Behaviour to Escape Local Minima for Swarms of Agents

An Emergent Wall Following Behaviour to Escape Local Minima for Swarms of Agents An Emergent Wall Followng Behavour to Escape Local Mnma for Swarms of Agents Mohamed H. Mabrouk and Coln R. McInnes Abstract Natural examples of emergent behavour, n groups due to nteractons among the

More information

Design and Optimization of Fuzzy Controller for Inverse Pendulum System Using Genetic Algorithm

Design and Optimization of Fuzzy Controller for Inverse Pendulum System Using Genetic Algorithm Desgn and Optmzaton of Fuzzy Controller for Inverse Pendulum System Usng Genetc Algorthm H. Mehraban A. Ashoor Unversty of Tehran Unversty of Tehran h.mehraban@ece.ut.ac.r a.ashoor@ece.ut.ac.r Abstract:

More information

Swarm Intelligence. SI systems are typically made up of a population of simple agents interacting locally with one another and with their environment.

Swarm Intelligence. SI systems are typically made up of a population of simple agents interacting locally with one another and with their environment. Swarm Intellgence Swarm ntellgence (SI) s an artfcal ntellgence technque based around the study of collectve behavor n decentralzed, selforganzed systems (Source: Wkpeda) SI systems are typcally made up

More information

Problem Set 9 Solutions

Problem Set 9 Solutions Desgn and Analyss of Algorthms May 4, 2015 Massachusetts Insttute of Technology 6.046J/18.410J Profs. Erk Demane, Srn Devadas, and Nancy Lynch Problem Set 9 Solutons Problem Set 9 Solutons Ths problem

More information

AP Physics 1 & 2 Summer Assignment

AP Physics 1 & 2 Summer Assignment AP Physcs 1 & 2 Summer Assgnment AP Physcs 1 requres an exceptonal profcency n algebra, trgonometry, and geometry. It was desgned by a select group of college professors and hgh school scence teachers

More information

(Online First)A Lattice Boltzmann Scheme for Diffusion Equation in Spherical Coordinate

(Online First)A Lattice Boltzmann Scheme for Diffusion Equation in Spherical Coordinate Internatonal Journal of Mathematcs and Systems Scence (018) Volume 1 do:10.494/jmss.v1.815 (Onlne Frst)A Lattce Boltzmann Scheme for Dffuson Equaton n Sphercal Coordnate Debabrata Datta 1 *, T K Pal 1

More information

Formation Control of Nonholonomic Multi-Vehicle Systems based on Virtual Structure

Formation Control of Nonholonomic Multi-Vehicle Systems based on Virtual Structure Proceedngs of the 7th World Congress The Internatonal Federaton of Automatc Control Seoul, Korea, July 6-, 8 Formaton Control of Nonholonomc Mult-Vehcle Systems based on Vrtual Structure Chka Yoshoka Toru

More information

Lecture Notes on Linear Regression

Lecture Notes on Linear Regression Lecture Notes on Lnear Regresson Feng L fl@sdueducn Shandong Unversty, Chna Lnear Regresson Problem In regresson problem, we am at predct a contnuous target value gven an nput feature vector We assume

More information

829. An adaptive method for inertia force identification in cantilever under moving mass

829. An adaptive method for inertia force identification in cantilever under moving mass 89. An adaptve method for nerta force dentfcaton n cantlever under movng mass Qang Chen 1, Mnzhuo Wang, Hao Yan 3, Haonan Ye 4, Guola Yang 5 1,, 3, 4 Department of Control and System Engneerng, Nanng Unversty,

More information

Resource Allocation with a Budget Constraint for Computing Independent Tasks in the Cloud

Resource Allocation with a Budget Constraint for Computing Independent Tasks in the Cloud Resource Allocaton wth a Budget Constrant for Computng Independent Tasks n the Cloud Wemng Sh and Bo Hong School of Electrcal and Computer Engneerng Georga Insttute of Technology, USA 2nd IEEE Internatonal

More information

COMPARISON OF SOME RELIABILITY CHARACTERISTICS BETWEEN REDUNDANT SYSTEMS REQUIRING SUPPORTING UNITS FOR THEIR OPERATIONS

COMPARISON OF SOME RELIABILITY CHARACTERISTICS BETWEEN REDUNDANT SYSTEMS REQUIRING SUPPORTING UNITS FOR THEIR OPERATIONS Avalable onlne at http://sck.org J. Math. Comput. Sc. 3 (3), No., 6-3 ISSN: 97-537 COMPARISON OF SOME RELIABILITY CHARACTERISTICS BETWEEN REDUNDANT SYSTEMS REQUIRING SUPPORTING UNITS FOR THEIR OPERATIONS

More information

Grover s Algorithm + Quantum Zeno Effect + Vaidman

Grover s Algorithm + Quantum Zeno Effect + Vaidman Grover s Algorthm + Quantum Zeno Effect + Vadman CS 294-2 Bomb 10/12/04 Fall 2004 Lecture 11 Grover s algorthm Recall that Grover s algorthm for searchng over a space of sze wors as follows: consder the

More information

DUE: WEDS FEB 21ST 2018

DUE: WEDS FEB 21ST 2018 HOMEWORK # 1: FINITE DIFFERENCES IN ONE DIMENSION DUE: WEDS FEB 21ST 2018 1. Theory Beam bendng s a classcal engneerng analyss. The tradtonal soluton technque makes smplfyng assumptons such as a constant

More information

Coupled Distributed Estimation and Control for Mobile Sensor Networks

Coupled Distributed Estimation and Control for Mobile Sensor Networks IEEE TRANSACTIONS ON AUTOMATIC CONTROL, VOL 57, NO 9, SEPTEMBER 1 1 Coupled Dstrbuted Estmaton and Control for Moble Sensor Networks Reza Olfat-Saber and Parsa Jalalkamal Abstract In ths paper, we ntroduce

More information

An Algorithm to Solve the Inverse Kinematics Problem of a Robotic Manipulator Based on Rotation Vectors

An Algorithm to Solve the Inverse Kinematics Problem of a Robotic Manipulator Based on Rotation Vectors An Algorthm to Solve the Inverse Knematcs Problem of a Robotc Manpulator Based on Rotaton Vectors Mohamad Z. Al-az*, Mazn Z. Othman**, and Baker B. Al-Bahr* *AL-Nahran Unversty, Computer Eng. Dep., Baghdad,

More information

This model contains two bonds per unit cell (one along the x-direction and the other along y). So we can rewrite the Hamiltonian as:

This model contains two bonds per unit cell (one along the x-direction and the other along y). So we can rewrite the Hamiltonian as: 1 Problem set #1 1.1. A one-band model on a square lattce Fg. 1 Consder a square lattce wth only nearest-neghbor hoppngs (as shown n the fgure above): H t, j a a j (1.1) where,j stands for nearest neghbors

More information

VQ widely used in coding speech, image, and video

VQ widely used in coding speech, image, and video at Scalar quantzers are specal cases of vector quantzers (VQ): they are constraned to look at one sample at a tme (memoryless) VQ does not have such constrant better RD perfomance expected Source codng

More information

First Law: A body at rest remains at rest, a body in motion continues to move at constant velocity, unless acted upon by an external force.

First Law: A body at rest remains at rest, a body in motion continues to move at constant velocity, unless acted upon by an external force. Secton 1. Dynamcs (Newton s Laws of Moton) Two approaches: 1) Gven all the forces actng on a body, predct the subsequent (changes n) moton. 2) Gven the (changes n) moton of a body, nfer what forces act

More information

Lecture 12: Discrete Laplacian

Lecture 12: Discrete Laplacian Lecture 12: Dscrete Laplacan Scrbe: Tanye Lu Our goal s to come up wth a dscrete verson of Laplacan operator for trangulated surfaces, so that we can use t n practce to solve related problems We are mostly

More information

Pop-Click Noise Detection Using Inter-Frame Correlation for Improved Portable Auditory Sensing

Pop-Click Noise Detection Using Inter-Frame Correlation for Improved Portable Auditory Sensing Advanced Scence and Technology Letters, pp.164-168 http://dx.do.org/10.14257/astl.2013 Pop-Clc Nose Detecton Usng Inter-Frame Correlaton for Improved Portable Audtory Sensng Dong Yun Lee, Kwang Myung Jeon,

More information

Formation Flight Control of Multi-UAV System with Communication Constraints

Formation Flight Control of Multi-UAV System with Communication Constraints do: 105028/jatmv8208 Formaton Flght Control of Mult-UAV System wth Communcaton Constrants Rubn Xue 1, Gaohua Ca 2 Abstract: Three dmensonal formaton control problem of mult-uav system wth communcaton constrants

More information

arxiv:cs.cv/ Jun 2000

arxiv:cs.cv/ Jun 2000 Correlaton over Decomposed Sgnals: A Non-Lnear Approach to Fast and Effectve Sequences Comparson Lucano da Fontoura Costa arxv:cs.cv/0006040 28 Jun 2000 Cybernetc Vson Research Group IFSC Unversty of São

More information

CSci 6974 and ECSE 6966 Math. Tech. for Vision, Graphics and Robotics Lecture 21, April 17, 2006 Estimating A Plane Homography

CSci 6974 and ECSE 6966 Math. Tech. for Vision, Graphics and Robotics Lecture 21, April 17, 2006 Estimating A Plane Homography CSc 6974 and ECSE 6966 Math. Tech. for Vson, Graphcs and Robotcs Lecture 21, Aprl 17, 2006 Estmatng A Plane Homography Overvew We contnue wth a dscusson of the major ssues, usng estmaton of plane projectve

More information

Chapter Newton s Method

Chapter Newton s Method Chapter 9. Newton s Method After readng ths chapter, you should be able to:. Understand how Newton s method s dfferent from the Golden Secton Search method. Understand how Newton s method works 3. Solve

More information

Appendix B: Resampling Algorithms

Appendix B: Resampling Algorithms 407 Appendx B: Resamplng Algorthms A common problem of all partcle flters s the degeneracy of weghts, whch conssts of the unbounded ncrease of the varance of the mportance weghts ω [ ] of the partcles

More information

Event-triggered Coordination for Formation Tracking Control in Constrained Space with Limited Communication

Event-triggered Coordination for Formation Tracking Control in Constrained Space with Limited Communication IEEE TRASACTIO O CYBERETICS Event-trggered Coordnaton for Formaton Trackng Control n Constraned Space wth Lmted Communcaton Xaome Lu, Shuzh Sam Ge, Fellow, IEEE, Cher-Hang Goh, Senor Member, IEEE, and

More information

Chapter - 2. Distribution System Power Flow Analysis

Chapter - 2. Distribution System Power Flow Analysis Chapter - 2 Dstrbuton System Power Flow Analyss CHAPTER - 2 Radal Dstrbuton System Load Flow 2.1 Introducton Load flow s an mportant tool [66] for analyzng electrcal power system network performance. Load

More information

COEFFICIENT DIAGRAM: A NOVEL TOOL IN POLYNOMIAL CONTROLLER DESIGN

COEFFICIENT DIAGRAM: A NOVEL TOOL IN POLYNOMIAL CONTROLLER DESIGN Int. J. Chem. Sc.: (4), 04, 645654 ISSN 097768X www.sadgurupublcatons.com COEFFICIENT DIAGRAM: A NOVEL TOOL IN POLYNOMIAL CONTROLLER DESIGN R. GOVINDARASU a, R. PARTHIBAN a and P. K. BHABA b* a Department

More information

EEE 241: Linear Systems

EEE 241: Linear Systems EEE : Lnear Systems Summary #: Backpropagaton BACKPROPAGATION The perceptron rule as well as the Wdrow Hoff learnng were desgned to tran sngle layer networks. They suffer from the same dsadvantage: they

More information

Week3, Chapter 4. Position and Displacement. Motion in Two Dimensions. Instantaneous Velocity. Average Velocity

Week3, Chapter 4. Position and Displacement. Motion in Two Dimensions. Instantaneous Velocity. Average Velocity Week3, Chapter 4 Moton n Two Dmensons Lecture Quz A partcle confned to moton along the x axs moves wth constant acceleraton from x =.0 m to x = 8.0 m durng a 1-s tme nterval. The velocty of the partcle

More information

Maximizing the number of nonnegative subsets

Maximizing the number of nonnegative subsets Maxmzng the number of nonnegatve subsets Noga Alon Hao Huang December 1, 213 Abstract Gven a set of n real numbers, f the sum of elements of every subset of sze larger than k s negatve, what s the maxmum

More information

Changing Topology and Communication Delays

Changing Topology and Communication Delays Prepared by F.L. Lews Updated: Saturday, February 3, 00 Changng Topology and Communcaton Delays Changng Topology The graph connectvty or topology may change over tme. Let G { G, G,, G M } wth M fnte be

More information

Multi-Robot Formation Control Based on Leader-Follower Optimized by the IGA

Multi-Robot Formation Control Based on Leader-Follower Optimized by the IGA IOSR Journal of Computer Engneerng (IOSR-JCE e-issn: 2278-0661,p-ISSN: 2278-8727, Volume 19, Issue 1, Ver. III (Jan.-Feb. 2017, PP 08-13 www.osrjournals.org Mult-Robot Formaton Control Based on Leader-Follower

More information

A Fast Computer Aided Design Method for Filters

A Fast Computer Aided Design Method for Filters 2017 Asa-Pacfc Engneerng and Technology Conference (APETC 2017) ISBN: 978-1-60595-443-1 A Fast Computer Aded Desgn Method for Flters Gang L ABSTRACT *Ths paper presents a fast computer aded desgn method

More information

Analysis of Robot Navigation Schemes using Rantzer s Dual Lyapunov Theorem

Analysis of Robot Navigation Schemes using Rantzer s Dual Lyapunov Theorem Analss of Robot Navgaton Schemes usng Rantzer s Dual Lapunov Theorem Dmos V. Dmarogonas and Karl H. Johansson Abstract When robots are drven b the negatve gradent of a potental feld that conssts of the

More information

Chapter 5. Solution of System of Linear Equations. Module No. 6. Solution of Inconsistent and Ill Conditioned Systems

Chapter 5. Solution of System of Linear Equations. Module No. 6. Solution of Inconsistent and Ill Conditioned Systems Numercal Analyss by Dr. Anta Pal Assstant Professor Department of Mathematcs Natonal Insttute of Technology Durgapur Durgapur-713209 emal: anta.bue@gmal.com 1 . Chapter 5 Soluton of System of Lnear Equatons

More information

Coverage Control with Information Decay in Dynamic Environments

Coverage Control with Information Decay in Dynamic Environments Coverage Control wth Informaton ecay n ynamc Envronments Nco Hübel S. Hrche A. Gusrald T. Hatanaka M. Fujta O. Sawodny Insttute for System ynamcs, Unverstät Stuttgart Fujta Laboratory, ept. of Mechancal

More information

COMPOSITE BEAM WITH WEAK SHEAR CONNECTION SUBJECTED TO THERMAL LOAD

COMPOSITE BEAM WITH WEAK SHEAR CONNECTION SUBJECTED TO THERMAL LOAD COMPOSITE BEAM WITH WEAK SHEAR CONNECTION SUBJECTED TO THERMAL LOAD Ákos Jósef Lengyel, István Ecsed Assstant Lecturer, Professor of Mechancs, Insttute of Appled Mechancs, Unversty of Mskolc, Mskolc-Egyetemváros,

More information

Why BP Works STAT 232B

Why BP Works STAT 232B Why BP Works STAT 232B Free Energes Helmholz & Gbbs Free Energes 1 Dstance between Probablstc Models - K-L dvergence b{ KL b{ p{ = b{ ln { } p{ Here, p{ s the eact ont prob. b{ s the appromaton, called

More information

PARTICIPATION FACTOR IN MODAL ANALYSIS OF POWER SYSTEMS STABILITY

PARTICIPATION FACTOR IN MODAL ANALYSIS OF POWER SYSTEMS STABILITY POZNAN UNIVE RSITY OF TE CHNOLOGY ACADE MIC JOURNALS No 86 Electrcal Engneerng 6 Volodymyr KONOVAL* Roman PRYTULA** PARTICIPATION FACTOR IN MODAL ANALYSIS OF POWER SYSTEMS STABILITY Ths paper provdes a

More information

A Swarm Aggregation Algorithm based on Local Interaction for Multi-Robot Systems with Actuator Saturations

A Swarm Aggregation Algorithm based on Local Interaction for Multi-Robot Systems with Actuator Saturations 01 IEEE/RSJ Internatonal Conference on Intellgent Robots and Systems October 7-1, 01. Vlamoura, Algarve, Portugal A Swarm Aggregaton Algorthm based on Local Interacton for Mult-Robot Systems wth Actuator

More information

Physics 5153 Classical Mechanics. Principle of Virtual Work-1

Physics 5153 Classical Mechanics. Principle of Virtual Work-1 P. Guterrez 1 Introducton Physcs 5153 Classcal Mechancs Prncple of Vrtual Work The frst varatonal prncple we encounter n mechancs s the prncple of vrtual work. It establshes the equlbrum condton of a mechancal

More information

Difference Equations

Difference Equations Dfference Equatons c Jan Vrbk 1 Bascs Suppose a sequence of numbers, say a 0,a 1,a,a 3,... s defned by a certan general relatonshp between, say, three consecutve values of the sequence, e.g. a + +3a +1

More information

MMA and GCMMA two methods for nonlinear optimization

MMA and GCMMA two methods for nonlinear optimization MMA and GCMMA two methods for nonlnear optmzaton Krster Svanberg Optmzaton and Systems Theory, KTH, Stockholm, Sweden. krlle@math.kth.se Ths note descrbes the algorthms used n the author s 2007 mplementatons

More information

Communication-efficient Distributed Solutions to a System of Linear Equations with Laplacian Sparse Structure

Communication-efficient Distributed Solutions to a System of Linear Equations with Laplacian Sparse Structure Communcaton-effcent Dstrbuted Solutons to a System of Lnear Equatons wth Laplacan Sparse Structure Peng Wang, Yuanq Gao, Nanpeng Yu, We Ren, Janmng Lan, and D Wu Abstract Two communcaton-effcent dstrbuted

More information

Operating conditions of a mine fan under conditions of variable resistance

Operating conditions of a mine fan under conditions of variable resistance Paper No. 11 ISMS 216 Operatng condtons of a mne fan under condtons of varable resstance Zhang Ynghua a, Chen L a, b, Huang Zhan a, *, Gao Yukun a a State Key Laboratory of Hgh-Effcent Mnng and Safety

More information

CHAPTER III Neural Networks as Associative Memory

CHAPTER III Neural Networks as Associative Memory CHAPTER III Neural Networs as Assocatve Memory Introducton One of the prmary functons of the bran s assocatve memory. We assocate the faces wth names, letters wth sounds, or we can recognze the people

More information

Fundamental loop-current method using virtual voltage sources technique for special cases

Fundamental loop-current method using virtual voltage sources technique for special cases Fundamental loop-current method usng vrtual voltage sources technque for specal cases George E. Chatzaraks, 1 Marna D. Tortorel 1 and Anastasos D. Tzolas 1 Electrcal and Electroncs Engneerng Departments,

More information

College of Computer & Information Science Fall 2009 Northeastern University 20 October 2009

College of Computer & Information Science Fall 2009 Northeastern University 20 October 2009 College of Computer & Informaton Scence Fall 2009 Northeastern Unversty 20 October 2009 CS7880: Algorthmc Power Tools Scrbe: Jan Wen and Laura Poplawsk Lecture Outlne: Prmal-dual schema Network Desgn:

More information

The Chaotic Robot Prediction by Neuro Fuzzy Algorithm (2) = θ (3) = ω. Asin. A v. Mana Tarjoman, Shaghayegh Zarei

The Chaotic Robot Prediction by Neuro Fuzzy Algorithm (2) = θ (3) = ω. Asin. A v. Mana Tarjoman, Shaghayegh Zarei The Chaotc Robot Predcton by Neuro Fuzzy Algorthm Mana Tarjoman, Shaghayegh Zare Abstract In ths paper an applcaton of the adaptve neurofuzzy nference system has been ntroduced to predct the behavor of

More information

Comparison of the Population Variance Estimators. of 2-Parameter Exponential Distribution Based on. Multiple Criteria Decision Making Method

Comparison of the Population Variance Estimators. of 2-Parameter Exponential Distribution Based on. Multiple Criteria Decision Making Method Appled Mathematcal Scences, Vol. 7, 0, no. 47, 07-0 HIARI Ltd, www.m-hkar.com Comparson of the Populaton Varance Estmators of -Parameter Exponental Dstrbuton Based on Multple Crtera Decson Makng Method

More information

Case A. P k = Ni ( 2L i k 1 ) + (# big cells) 10d 2 P k.

Case A. P k = Ni ( 2L i k 1 ) + (# big cells) 10d 2 P k. THE CELLULAR METHOD In ths lecture, we ntroduce the cellular method as an approach to ncdence geometry theorems lke the Szemeréd-Trotter theorem. The method was ntroduced n the paper Combnatoral complexty

More information

Tracking with Kalman Filter

Tracking with Kalman Filter Trackng wth Kalman Flter Scott T. Acton Vrgna Image and Vdeo Analyss (VIVA), Charles L. Brown Department of Electrcal and Computer Engneerng Department of Bomedcal Engneerng Unversty of Vrgna, Charlottesvlle,

More information

Department of Statistics University of Toronto STA305H1S / 1004 HS Design and Analysis of Experiments Term Test - Winter Solution

Department of Statistics University of Toronto STA305H1S / 1004 HS Design and Analysis of Experiments Term Test - Winter Solution Department of Statstcs Unversty of Toronto STA35HS / HS Desgn and Analyss of Experments Term Test - Wnter - Soluton February, Last Name: Frst Name: Student Number: Instructons: Tme: hours. Ads: a non-programmable

More information

Collision-Free Path and Trajectory Planning Algorithm for Multiple-Vehicle Systems

Collision-Free Path and Trajectory Planning Algorithm for Multiple-Vehicle Systems Collson-Free Path and Trajectory Plannng Algorthm for Multple-Vehcle Systems Anugrah K. Pamosoaj, and Keum-Sh Hong, Senor Member, IEEE Abstract An algorthm to generate collson-free paths and trajectores

More information

Irregular vibrations in multi-mass discrete-continuous systems torsionally deformed

Irregular vibrations in multi-mass discrete-continuous systems torsionally deformed (2) 4 48 Irregular vbratons n mult-mass dscrete-contnuous systems torsonally deformed Abstract In the paper rregular vbratons of dscrete-contnuous systems consstng of an arbtrary number rgd bodes connected

More information

12. The Hamilton-Jacobi Equation Michael Fowler

12. The Hamilton-Jacobi Equation Michael Fowler 1. The Hamlton-Jacob Equaton Mchael Fowler Back to Confguraton Space We ve establshed that the acton, regarded as a functon of ts coordnate endponts and tme, satsfes ( ) ( ) S q, t / t+ H qpt,, = 0, and

More information

NP-Completeness : Proofs

NP-Completeness : Proofs NP-Completeness : Proofs Proof Methods A method to show a decson problem Π NP-complete s as follows. (1) Show Π NP. (2) Choose an NP-complete problem Π. (3) Show Π Π. A method to show an optmzaton problem

More information

Gravitational Acceleration: A case of constant acceleration (approx. 2 hr.) (6/7/11)

Gravitational Acceleration: A case of constant acceleration (approx. 2 hr.) (6/7/11) Gravtatonal Acceleraton: A case of constant acceleraton (approx. hr.) (6/7/11) Introducton The gravtatonal force s one of the fundamental forces of nature. Under the nfluence of ths force all objects havng

More information

Autonomous cooperative driving: a velocity-based negotiation approach for intersection crossing

Autonomous cooperative driving: a velocity-based negotiation approach for intersection crossing Autonomous cooperatve drvng: a velocty-based negotaton approach for ntersecton crossng Gabrel Rodrgues de Campos, Paolo Falcone and Jonas Sjöberg Abstract In ths artcle, a scenaro where several vehcles

More information

A PROBABILITY-DRIVEN SEARCH ALGORITHM FOR SOLVING MULTI-OBJECTIVE OPTIMIZATION PROBLEMS

A PROBABILITY-DRIVEN SEARCH ALGORITHM FOR SOLVING MULTI-OBJECTIVE OPTIMIZATION PROBLEMS HCMC Unversty of Pedagogy Thong Nguyen Huu et al. A PROBABILITY-DRIVEN SEARCH ALGORITHM FOR SOLVING MULTI-OBJECTIVE OPTIMIZATION PROBLEMS Thong Nguyen Huu and Hao Tran Van Department of mathematcs-nformaton,

More information

EPR Paradox and the Physical Meaning of an Experiment in Quantum Mechanics. Vesselin C. Noninski

EPR Paradox and the Physical Meaning of an Experiment in Quantum Mechanics. Vesselin C. Noninski EPR Paradox and the Physcal Meanng of an Experment n Quantum Mechancs Vesseln C Nonnsk vesselnnonnsk@verzonnet Abstract It s shown that there s one purely determnstc outcome when measurement s made on

More information

Nodal analysis of finite square resistive grids and the teaching effectiveness of students projects

Nodal analysis of finite square resistive grids and the teaching effectiveness of students projects 2 nd World Conference on Technology and Engneerng Educaton 2 WIETE Lublana Slovena 5-8 September 2 Nodal analyss of fnte square resstve grds and the teachng effectveness of students proects P. Zegarmstrz

More information

Lab 2e Thermal System Response and Effective Heat Transfer Coefficient

Lab 2e Thermal System Response and Effective Heat Transfer Coefficient 58:080 Expermental Engneerng 1 OBJECTIVE Lab 2e Thermal System Response and Effectve Heat Transfer Coeffcent Warnng: though the experment has educatonal objectves (to learn about bolng heat transfer, etc.),

More information

Supplemental document

Supplemental document Electronc Supplementary Materal (ESI) for Physcal Chemstry Chemcal Physcs. Ths journal s the Owner Socetes 01 Supplemental document Behnam Nkoobakht School of Chemstry, The Unversty of Sydney, Sydney,

More information

Feature Selection: Part 1

Feature Selection: Part 1 CSE 546: Machne Learnng Lecture 5 Feature Selecton: Part 1 Instructor: Sham Kakade 1 Regresson n the hgh dmensonal settng How do we learn when the number of features d s greater than the sample sze n?

More information

Simulation for Pedestrian Dynamics by Real-Coded Cellular Automata (RCA)

Simulation for Pedestrian Dynamics by Real-Coded Cellular Automata (RCA) Smulaton for Pedestran Dynamcs by Real-Coded Cellular Automata (RCA) Kazuhro Yamamoto 1*, Satosh Kokubo 1, Katsuhro Nshnar 2 1 Dep. Mechancal Scence and Engneerng, Nagoya Unversty, Japan * kazuhro@mech.nagoya-u.ac.jp

More information

Preserving Strong Connectivity in Directed Proximity Graphs

Preserving Strong Connectivity in Directed Proximity Graphs Preservng Strong Connectvty n Drected Proxmty Graphs Hasan A. Poonawala, and Mark W. Spong, Fellow, IEEE Abstract Ths paper proposes a method to mantan the strong connectvty property of a moble robot ad

More information

A Robust Method for Calculating the Correlation Coefficient

A Robust Method for Calculating the Correlation Coefficient A Robust Method for Calculatng the Correlaton Coeffcent E.B. Nven and C. V. Deutsch Relatonshps between prmary and secondary data are frequently quantfed usng the correlaton coeffcent; however, the tradtonal

More information

Section 8.3 Polar Form of Complex Numbers

Section 8.3 Polar Form of Complex Numbers 80 Chapter 8 Secton 8 Polar Form of Complex Numbers From prevous classes, you may have encountered magnary numbers the square roots of negatve numbers and, more generally, complex numbers whch are the

More information

Design and Analysis of Landing Gear Mechanic Structure for the Mine Rescue Carrier Robot

Design and Analysis of Landing Gear Mechanic Structure for the Mine Rescue Carrier Robot Sensors & Transducers 214 by IFSA Publshng, S. L. http://www.sensorsportal.com Desgn and Analyss of Landng Gear Mechanc Structure for the Mne Rescue Carrer Robot We Juan, Wu Ja-Long X an Unversty of Scence

More information

Dynamic Systems on Graphs

Dynamic Systems on Graphs Prepared by F.L. Lews Updated: Saturday, February 06, 200 Dynamc Systems on Graphs Control Graphs and Consensus A network s a set of nodes that collaborates to acheve what each cannot acheve alone. A network,

More information

Pivot-Wheel Drive Crab with a Twist! Clem McKown Team November-2009 (eq 1 edited 29-March-2010)

Pivot-Wheel Drive Crab with a Twist! Clem McKown Team November-2009 (eq 1 edited 29-March-2010) Pvot-Wheel Drve Crab wth a Twst! Clem McKown Team 1640 13-November-2009 (eq 1 edted 29-March-2010) 4-Wheel Independent Pvot-Wheel Drve descrbes a 4wd drve-tran n whch each of the (4) wheels are ndependently

More information

Chapter 8 Indicator Variables

Chapter 8 Indicator Variables Chapter 8 Indcator Varables In general, e explanatory varables n any regresson analyss are assumed to be quanttatve n nature. For example, e varables lke temperature, dstance, age etc. are quanttatve n

More information

χ x B E (c) Figure 2.1.1: (a) a material particle in a body, (b) a place in space, (c) a configuration of the body

χ x B E (c) Figure 2.1.1: (a) a material particle in a body, (b) a place in space, (c) a configuration of the body Secton.. Moton.. The Materal Body and Moton hyscal materals n the real world are modeled usng an abstract mathematcal entty called a body. Ths body conssts of an nfnte number of materal partcles. Shown

More information

Amplification and Relaxation of Electron Spin Polarization in Semiconductor Devices

Amplification and Relaxation of Electron Spin Polarization in Semiconductor Devices Amplfcaton and Relaxaton of Electron Spn Polarzaton n Semconductor Devces Yury V. Pershn and Vladmr Prvman Center for Quantum Devce Technology, Clarkson Unversty, Potsdam, New York 13699-570, USA Spn Relaxaton

More information

Queueing Networks II Network Performance

Queueing Networks II Network Performance Queueng Networks II Network Performance Davd Tpper Assocate Professor Graduate Telecommuncatons and Networkng Program Unversty of Pttsburgh Sldes 6 Networks of Queues Many communcaton systems must be modeled

More information