On Event-Triggered and Self-Triggered Control over Sensor/Actuator Networks

Size: px
Start display at page:

Download "On Event-Triggered and Self-Triggered Control over Sensor/Actuator Networks"

Transcription

1 Proceeings of the th IEEE Conference on Decision an Control Cancun, Mexico, Dec. 9-11, 8 TuA1. On Event-Triggere an Self-Triggere Control over Sensor/Actuator Networks Manuel Mazo Jr. an Paulo Tabuaa Abstract Event-triggere an self-triggere control have been recently propose as an alternative to the more traitional perioic execution of control tasks. The possibility of reucing the number of executions while guaranteeing esire levels of performance makes event-triggere an self-triggere control very appealing in the context of sensor/actuator networks. In this setting, reucing the number of times that a feeback control law is execute implies a reuction in transmissions an thus a reuction in energy expenitures. In this paper we introuce two novel istribute implementations of event-triggere an self-triggere policies over sensor/actuator networks an iscuss their performance in terms of energy expeniture. I. INTRODUCTION Sensor networks research has extensively ealt with the extraction of information from the physical worl. Many of the evelope applications concentrate on how to obtain this information for posterior off-line analysis [1], [] or on-line processing of this information for tracking [], [], istribute optimization [5] or mapping []. In all of these applications there is a common esire for small power consumption, which woul exten the life span of the network. Many of the approaches use to reuce the power consumption concentrate on the communication requirements, as the works on compressive sensing [], network throughput optimization [8], [9], message-passing algorithms [1] or sleep-scheuling of the noes [11]. Still, most of these stuies are performe only for on- or off-line analysis on the information gathere. We aress the problem of minimizing energy consumption for applications in which actuation plays a major role, namely control applications. In this context, energy expenitures can be irectly relate to the frequency at which measurements are being taken an transmitte through the network. Although the choice of the sampling frequency is a problem as ol as the use of microprocessors for control, this question never receive a efinitive answer an engineers still rely on rules of thumb [1], [1], [1] such as sampling with a frequency times the system banwith. Moreover, the choice of perioic sampling is not even a natural one since the behavior of a control system can be quite ifferent in ifferent regions of the state space. In this paper, we show how to minimize energy consumption by resorting to event-triggere an self-triggere This work has been partially fune by the Spanish Ministry of Science an Eucation/UCLA fellowship LA- an by the NSF CSR-EHS 15 grant. M. Mazo Jr. an P. Tabuaa are with the Department of Electrical Engineering, University of California at Los Angeles, Los Angeles, CA , USA mmazo,tabuaa@ee.ucla.eu. sampling strategies over sensor/actuator networks (SAN). The event-triggere iea has been previously explore in the literature [15]. It has been shown in [1] that by sampling an computing the controller only when a certain threshol conition on the state is violate, one can sample less frequently than with a perioic scheme while guaranteeing the same performance. In [1] an [18] the self-triggere sampling paraigm was explore, for linear an non-linear systems respectively, in an attempt to eliminate the requirement of continuous state measurement. Self-triggere strategies compute the next sampling time using the last state measurement an thus o not require knowlege of the current state. The contribution of this paper is to show how the techniques introuce in [1], an later refine in [19], [1] for H controllers, can be implemente over sensor-actuator networks to consierably reuce the number of network transmissions. We propose an event-triggere strategy in which each noe uses its local information to etermine when to make a transmission an a self-triggere strategy in which the actuator noe etermines for how long shoul the sensing noes sleep before collecting an transmitting fresh measurements. The paper is organize in the following way: Section II introuces the notation an states the problem. Then we procee to escribe the propose ecentralize self-triggere algorithm in Section III. Next we propose a istribute eventtriggere protocol in Section IV, followe by Section V where we formalize how to obtain the times between upates in the previously escribe algorithms. The propose techniques are illustrate with two examples in Section VI. The paper conclues with a brief iscussion in Section VII. II. NOTATION AND PROBLEM STATEMENT We consier a linear control system: ẋ = Ax + Bu, x R n, u R m (1) stabilize by a linear controller: u = Kx () In what follows, we assume that K was obtaine through a LQR esign an consier its implementation over a sensor/actuator network. Therefore, K is given by K = B T P where P is the solution of the matrix Riccati equation: A T P + P A Q + L = () with Q = P BB T P an for some L an P = P T >. Furthermore, we assume m=1, for simplicity of presentation, an also the existence of a istinguishe noe collocate /8/$5. 8 IEEE 5

2 th IEEE CDC, Cancun, Mexico, Dec. 9-11, 8 TuA1. with the actuator calle the actuator noe. The remaining noes are sensor noes an further escribe below. The control signal is obtaine from zero-orer sample an hol measurements so that: u(t) = Kx(t k ), t [t k, t k+1 ) where t k an t k+1 are two consecutive sampling instants. For the sake of reaability, we rop the explicit epenence of t but enote with the subinex k those quantities that are constant an equal to the value at the upate time t k, i.e. x k = x(t k ). With this kin of sample feeback the close loop is escribe by: ẋ = Ax + BKx k = (A + BK)x + BKe where e represents a measurement error efine by: t [t k, t k+1 ) = e(t) = x(t k ) x(t) Treating the measurement error e as a new variable we can rewrite the close loop ynamics as: [ ] [ ] [ ] ẋ A + BK BK x = ė A BK BK e since ė = ẋ. Moreover, the matrix P, obtaine as the solution of (), efines a quaratic Lyapunov function V = x T P x whose erivative along () satisfies: V = x T Lx + e T Qe u T k u k The esire performance for the close loop system can be escribe by a matrix S > for which the inequality: V x T Sx () hols. The matrix S woul then efine the esire rate of ecay for V. In orer to enforce () we nee to select the sampling instants t k so that x T Lx + e T Qe u T k u k x T Sx. This can be achieve by sampling when (5) hols. x T (L S)x + e T Qe u T k u k = (5) At t = t k the execution of the controller reners x k = x an e = x k x =, forcing x T k (Q+L S)x k. Now we can appreciate how (Q + L S) captures the trae-off between sampling an performance. In a typical esign one woul pick L to esign the continuous optimal controller, Q will be etermine by the LQR esign, an we will get to pick again S. The closer S an L are, the better the performance of the system will be (closer to the prescribe continuous esign) at the expense of more frequent sampling. In a practical setting the measurements x k will actually be receive after a certain elay which shoul be incorporate into the efinition of e. For the sake of simplicity we ignore this elay by assuming = although we can incorporate the elay in the propose framework by appropriately selecting S in the rule () by following the approach escribe in [1]. We are intereste in the use of event-triggere an selftriggere control techniques as they require fewer measurements than perioic techniques. In particular, in the context of SAN s, this also means that less transmissions are require for control an thus less energy usage. We will assume without loss of generality that each sensor measures one entry of the full state of the system. With all of the above in min, the problem we solve in this paper is: Problem.1: Given a Linear Time Invariant system as in (1), a controller () renering the close-loop system asymptotically stable, esign a istribute algorithm enforcing the controller upate rule () over a SAN. We can now see that the funamental ifficulty when implementing an event-triggere or self-triggere strategy in a istribute-sensing setting is that the triggering conition () has to be evaluate continuously an it epens on full state information. A continuous centralization of local measurements to check () woul lea to extensive communication an unacceptably high energy consumption, thus justifying a istribute esign. In aition to a reuction in energy consumption, we will also require the istribute algorithms to be energy-balance, i.e. all the sensing noes shoul consume energy at the same pace, since as soon as one sensing noe exhausts its energy reserves the stability an performance of the close loop system becomes seriously compromise. In what follows will enote the usual Eucliean norm of a vector an σ m,m ( ) the minimum (m) or maximum (M) singular value of a given matrix. III. A SELF-TRIGGERED PROTOCOL The first istribute implementation follows the selftriggere paraigm for control [], [1], [18]. Accoring to this paraigm, in aition to use the current state to compute the controller, the current state is also use to compute the next execution time of the controller. In this section we escribe a particular protocol to implement this paraigm over a sensor/actuator network. A. Algorithm escription We start by assuming the availability of a routing tree with root at the actuation noe. The construction of a routing tree is typically one uring the iscovery phase of the network by resorting to several efficient algorithms available in the literature [1]. We also assume that the actuator noe has access to larger energy reserves since it rives one or several actuators that are typically more power consuming. Therefore, we assume that the actuation noe has the ability to broacast messages to all noes without seriously reucing its energy reserves. The propose ecentralize algorithm for self-triggere control consists of two phases: A measurement collection phase in which the network computes in a istribute manner, as explaine later in this section, two scalars: u k = Kx k an x k. The actuator noe uses this information to upate the actuator an to compute the next time instant at which new samples shoul be collecte an transmitte. A broacast phase uring which the actuator noe broacasts the next upate time to all the sensing noes.

3 th IEEE CDC, Cancun, Mexico, Dec. 9-11, 8 TuA1. In practice, in orer to avoi synchronization issues one woul rather sen t k+1 t k instea of t k+1. The sensing noes woul then sleep for t k+1 t k units of time before starting the next measurement collection phase. These t k+1 t k times will be compute in Section V. In any implementation of a control system the control signal u has to be elivere to the controller. In aition to u k we will also nee to transmit aitional information in orer to preict the sampling times. For the techniques that we will present x k is enough; nevertheless, more information will help to increase the time between upates. The istribute computation of u k = Kx k an x k = n 1 x i (t k) is base on the fact that both quantities are sums of quantities locally available to the noes of the network. The computation of these quantities can then be one by a wave algorithm for trees (Tree Wave-Algorithm) [1] in which each noe collects its local quantity (K i x i k an (xi k ) respectively), as it to the quantities that it receives from its chilren an transmits the resulting sum to its parent. Having the root at the actuation noe ensures that the computation ens at the actuation noe. The algorithm also ensures that each noe transmits packets of the same size. The number of transmissions are also limite to one per noe, ue to the Tree Wave-Algorithm implementation. Briefly, the Tree Wave-Algorithm limits the communication by passing to the parent noe a token or signal only when the noe has receive one from each of the links to its chilren, an it has locally generate its own token. In this way when the root receives a token, the root noe knows that all noes in the network generate locally a token. If the maximum elay introuce by the measuring collection phase can be compute, it can be regare as a computation elay an accommoate by moifying the triggering rule as in [1]. B. Energy utilization In orer to etermine the energy consume by the propose protocol, we nee to take into account the power consume in transmitting ata, in listening for transmissions, an in computing. In the case of a self-triggere strategy the noes only perform the computations neee in the wave algorithm computing u k an x k. We will enote this computing cost by D 1. It is a fixe unit of energy consume per upate. The raio moule can also be kept asleep most of the time as there is no nee for communication between upates, therefore the listening cost coul be neglecte. As for the power consume in transmitting information, the noes on the self-triggere protocol only sen two scalars per upate. Thus, we represent this cost by two parameters: C 1 an C. The first parameter escribes the cost associate to the packet overhea transmission (inepenent of the payloa) while the secon parameter escribes the cost per transmitte scalar. Therefore, in the self-triggere case there woul be an energy cost of C 1 + C, an the total energy consume at a noe at any given point in time will be given by: J 1 (t) = (C 1 + C + D 1 ) () k Γ(t) where Γ(t) = {k : t k t}. IV. AN EVENT-TRIGGERED PROTOCOL A ifferent strategy consists in checking in a istribute manner the conition impose on the ecay of the Lyapunov equation. This leas to a istribute computation of the controller upate times. Our goal is to reuce the consumption at the noes of the network, on which communication has the biggest impact. Therefore we cannot exchange information frequently between noes. Instea we propose to have the noes estimate in a conservative way when the triggering conition stops to hol. Each of the noes will be making use of its local measurements in orer to perform this estimation. When all the noes agree in that the triggering conition has been violate a new controller upate is force. A. Algorithm escription Once more we will nee to assume the existence of a routing tree with root at the actuation noe visiting all the noes in the network. The event-triggere algorithm can be ivie in several phases: Phase 1: The actuation noe broacasts a request for new measurements. Phase : The sensing noes take their local measurement an compute their part of u k an x k an forwar them up the routing tree (towars the root). Phase : The actuation noe finishes the computation of u k an x k an broacasts those quantities to the rest of the noes. Phase : The sensing noes use u k an x k to upate their local estimators accoringly. When a noe s local triggering conition stops to hol it will generate a local token, starting the computation of the Tree Wave- Algorithm, explaine in a previous section. When the actuation noe receives tokens from all of its chilren noes it will broacast a request for new measurements an the network woul be back in Phase 1. The local triggering (generation of these tokens) will be explaine in etail in Section V. As the measurements from the noes are synchronize, any elay introuce in phases an can be regare as a computation elay an again ealt with moifying the triggering rule as in [1]. Note also that the performance of neither the self-triggere nor the event-triggere protocol is affecte by the topology of the network if no elays are taken into account. However, the topology will have an effect on the maximum elay present in a practical implementation, an thus on its performance. B. Energy utilization The event-triggere protocol requires the noes to continuously compute estimates, an listen to possible communications from their chilren. Therefore, we have a constant

4 th IEEE CDC, Cancun, Mexico, Dec. 9-11, 8 TuA1. power consumption ue to listening an computation, which we will enote by l an respectively. In the event-triggere protocol there are two transmissions per noe per upate. The first one is use just to ecie the next upate time an contains no payloa, therefore incurring on a cost of C 1. The secon transmission is use to compute the control signal u k an x k, therefore having two scalars as payloa an incurring on a cost of C 1 +C. Finally the total energy consumption at any noe at any point in time will be given by the following equation: J (t) = (C 1 + C ) + (l + ) t () k Γ(t) where once more Γ(t) = {k : t k t}. V. HOW TO COMPUTE SLEEPING TIMES We will nee to relax conition () in orer to make it possible to check in a ecentralize way. In this section we will present these relaxations an how to take avantage of them to perform a istribute or ecentralize computation of the times between upates of the controller. A. Self-triggere control One of these possible relaxations provies the triggering conition: e T (Q + L S)e 1 σ m(l S) x k + u T k u k (8) which implies the inequalities: e T (Q + L S)e 1 xt k (L S)x k + u T k u k V x T Sx with α = M 1 AM 1 (1) β = min{β 1, β } (11) β 1 = M 1 (A + BK) xk ( (1) β = M 1 (A + BK) x k + (1) M 1 Buk + u T k B T M 1 A xk Our boun estimator woul therefore be ɛ = αɛ+β, which by the Comparison Lemma [], woul provie with ɛ(t) e(t) M. Integrating ɛ an taking into account ɛ(t k ) = leas to: ɛ(t) = β ( ) e α(t tk) 1 (1) α Therefore the next upate time will be given by: ) 1 ɛ (t k+1 ) = σ m(l S) x k + u T k u k (15) which can be solve analytically: t k+1 = t k + 1 ( α log 1 + ακ ) k (1) β σm (L S) κ k = x k + u T k u k (1) Let us summarize these results in the following proposition: Proposition 5.1: The istribute implementation of a LQR controller (), esigne to stabilize the linear system (1), efine by the algorithm escribe in Section III with upate times given by (1) guarantees stability of the close loop system an enforces V x T Sx. Moreover, the algorithm is energy-balance. also hol, where we use: x T (L S)x 1 xt k (L S)x k e T (L S)e The right-han-sie of equation (8) is a linear combination of the constants u k an x k obtaine from a istribute computation involving only the last measurements taken by the sensing noes. In orer to check the triggering conition on-line we will nee to estimate an upper boun for the evolution of e T (Q + L S)e. Provie that L an S where esigne so that (Q + L S) > we can fin a ynamic estimator ɛ satisfying e(t) M ɛ(t) an bouning above e M = M 1 e = Q + L Se : t e M ė M M 1 AM 1 e M + (A+BK)x k M Which now can be moifie to be use as a boun estimator epening only on quantities computable over the network: t e M α e M + β(u k, x k ) (9) 8 B. Event-triggere control A ifferent alternative consists of esigning estimators in which local measurements can be irectly injecte. In this way the triggering conitions can be locally checke. In orer to be able to obtain such estimators we also nee to relax the triggering conition () in the local noes. σ M (Q) e ( σ m (L S) x + u T ) k u k (18) V x T Sx Now, as in the esign of the self-triggere estimator, we can obtain estimators with irect injection of local measurements for e an x. Let us first partition an reorer the state-space as e i = [ e i, ê i] T, with ei enoting the local error variable at noe i, an ê i the remaining entries of the error vector. With this new escription of the error vector at noe i, we can rewrite the ynamics for e as t [ ei ê i ] = [ A1 A ] [ ei ê i ] A A [ ] [ A1 A xi (t k ) A A ˆx i (t k ) ] [ ] B1 u B k

5 th IEEE CDC, Cancun, Mexico, Dec. 9-11, 8 TuA1. Then, t ê i A e i + A ê i B u k [A A ]xk i A ê i + ( A e i B u k + [A A ] x k + (A e i B u k ) T [A A ] x k ) 1 an hence we can use as boun estimator: e T e e i + (ε i ) (19) t εi = A ε i + ` A e i B u k + [A A ] x k + (A e i B u k ) T [A A ] x k 1 () In an analogous way we can erive a lower boun estimator for x resulting in: x T x x i + (χ i ) (1) t χi = A χ i A x i + B u k () Therefore, local triggering conitions of the form σ M (Q) e i + (ε i ) σ m(l S) x i + (χ i ) + u T k u k () coul be use to ecie the upate times. The istribute algorithm propose in section IV-A will enforce a new upate only when all noes have their local rules violate. Therefore, the istribute algorithm will ensure that the controller follows the noe with least conservative estimates, an thus the system waits as long as possible before triggering a new upate. Now we can summarize the event-triggere protocol in the following proposition: Proposition 5.: The istribute implementation of a LQR controller (), esigne to stabilize the linear system (1), efine by the algorithm escribe in Section IV with upate times given by (),(), an () guarantees stability of the close loop system an enforces V x T Sx. Moreover, the algorithm is energy-balance. VI. EXAMPLE We illustrate these techniques in two small examples. Let us start with the system efine by A = , B = 1 1 5, xo = 1 The controller was esigne solving the Riccati equation (), with L =.5I an setting K = B T P. The S matrix was set to S =.I. For comparison we inclue plots of simulations using the centralize event-triggere rule e T Qe u T k u k x T (L S)x, which oes not involve any approximation or relaxation from (). This protocol woul be giving us the exact times when the controller nees to be upate. Obviously our two protocols for networks will give us more upates, but not necessarily always smaller times. In figure 1 we can see how the three strategies achieve the esire performance as measure by the ecay of the 5 x T Px Centralize Event Triggere Distribute Event Triggere Distribute Self Triggere Fig. 1. Lyapunov function ecay for centralize event-triggere, istribute event-triggere an ecentralize self-triggere. First example. t k+1 t k [s] Centralize Event Triggere Distribute Event Triggere Decentralize Self Triggere Fig.. Inter-sample times for centralize event-triggere, istribute eventtriggere an ecentralize self-triggere strategies for the first example. Lyapunov function. But looking at figure, while the times obtaine with the self-triggere strategy are much shorter than the centralize event-triggere controller, the istribute event-triggere strategy prouces times much closer to those from the centralize controller. This effect is explaine by the fact that in an event-triggere protocol measurements from the plant at a local level are taken continuously, while the self-triggere protocol is completely open loop between upates. Finally we present a comparison of the energy consumption evolution at a noe (an therefore at all of them as our algorithms are balance). The values for the ifferent costs use in the energy computation were: C 1 = 8.mJ, C =.mj, l = mw, = mw an D 1 = mj negligible. These values are approximations obtaine from the actual power consumption values for a MicaZ [] using ZigBee [] for wireless communication. Figure epicts the energy consumption using the self-triggere an the eventtriggere protocols, as given by () an (). Observe how the istribute event-triggere algorithm consumes less energy than the self-triggere. In this example the event-triggere strategy becomes more energy-efficient because of the great ifference on number of upates between event an selftriggere. But in general that oes not have to be the case. Even if the event-triggere algorithm prouces fewer upates than the self-triggere, the first algorithm coul consume more energy. The next example tries to portray this situation. 9

6 th IEEE CDC, Cancun, Mexico, Dec. 9-11, 8 TuA1. J 1, [mj] 1 x Distribute Event Triggere Distribute Self Triggere Hybri Strategy Fig.. Energy consumption for istribute event-triggere an ecentralize self-triggere strategies for the first example. J 1, [mj] Distribute Event Triggere Distribute Self Triggere Hybri Strategy Fig.. Energy consumption for istribute event-triggere an ecentralize self-triggere strategies for the secon example. The system is escribe by: A = , B = x o = [ ] T , an we use again L =.5I an S =.I. In figure we can appreciate how the self-triggere protocol again prouces more frequent upates, but if we look at the energy consumption we can see how the istribute event-triggere algorithm consumes more energy all the time. In this example the evolution of the system is quite slow, an therefore the listening an computing energy plays a bigger role, making the istribute event-triggere strategy less efficient. A hybri strategy in which the event-triggere protocol also uses the self-triggere preicte times to keep the raio-moule asleep is also presente in both figure an figure. This algorithm improves the performance of the event-triggere algorithm but oes not always outperform the self-triggere algorithm. VII. DISCUSSION We have propose two istribute implementations for controllers over SAN s base on the event-triggere an selftriggere control paraigms. The use of event-triggere an self-triggere control is justifie in terms of the reuction in energy consumption obtaine by reucing the number of messages exchange over the network while achieving the prescribe control performance. Both strategies were illustrate through two examples which show that no algorithm outperforms the other in terms of energy expeniture. A hybri strategy benefiting from both approaches was teste. This hybri protocol always outperforms the event-triggere approach, but that is not the case with the self-triggere protocol. The selection of the most aequate protocol for a given control system is left for future research. Note that the results obtaine strongly epen on the particular estimators selecte. We are currently investigating how to improve the performance of the propose estimators. REFERENCES [1] J. Fisher, T. Harmon, an W. Kaiser, Multiscale river hyraulic an water quality observations combining stationary an mobile sensor network noes, in American Geophysical Union Joint Assembly Annual Spring Meeting, Baltimore, MD, May -. [] L. Giro an M. Roch, An overview of the use of remote embee sensors for auio acquisition an processing. in In the Proceeings of the IEEE International Symposium on Multimeia (ISM), San Diego, CA, December. [] S. Oh an S. Sastry, Tracking on a graph, in Information Processing in Sensor Networks (IPSN), 5. [] B. Sinopoli, C. Sharp, L. Schenato, S. Shaffert, an S. Sastry, Distribute control applications within sensor networks, in Proceeings of the IEEE,Special Issue on Sensor Networks an Applications, 5. [5] M. Rabbat an R. Nowak, Distribute optimization in sensor networks, in Information Processing in Sensor Networks (IPSN),. [] J. Djugash, S. Singh, an B. Grocholsky, Decentralize mapping of robot-aie sensor networks, in IEEE International Conference on Robotics an Automation, 8. [] W. Bajwa, J. Haupt, A. Sayee, an R. Nowak, Compressive wireless sensing, in Information Processing in Sensor Networks (IPSN),. [8] A. Chakrabarti, A. Sabharwal, an B. Aazhang, Multi-hop communication in orer-optimal for homogeneous sensor networks, in Information Processing in Sensor Networks (IPSN),. [9] H. Le, D. Henriksson, an T. Abelzaher, A control therory approach to throughput optimization in multi-channel collection sensor networks, in Information Processing in Sensor Networks (IPSN),. [1] M. Paskin, C. Guestrin, an J. McFaen, A robust architecture for istribute inference in sensor networks, in Information Processing in Sensor Networks (IPSN), 5. [11] R. Subramanian an F. Fekri, Sleep scheuling an lifetime maximization in sensor networks: Funamental limits an optimal solutions, in Information Processing in Sensor Networks (IPSN),. [1] G. Franklin, Rational rate, in IEEE Control Systems Magazine,. [1] G. Goowin, S. Graebe, an M. Salgao, Control System Design. Prentice Hall, 1. [1] C. Houpis an G. B. Lamont, Digital Control Systems. McGraw-Hill Higher Eucation, 198. [15] K. Åström an B. Bernharsson, Comparison of Riemann an Lebesgue sampling for first orer stochastic systems, Proceeings of the 1st IEEE Conference on Decision an Control, vol.,. [1] P. Tabuaa, Event-triggere real-time scheuling of stabilizing control tasks, in IEEE Transactions on Automatic Control, vol. 5(9),. [1] X. Wang an M. D. Lemmon, State base self-triggere feeback control systems with l stability, in 1th IFAC Worl Congress, 8. [18] A. Anta an P. Tabuaa, Self-triggere stabilization of homogeneous control systems, in American Control Conference, 8. [19] X. Wang an M. D. Lemmon, Event-triggere broacasting across istribute networke control systems, in American Control Conference, 8. [] M. Velasco, J. Fuertes, an P. Marti, The self triggere task moel for real-time control systems, Work in Progress Proceeings of the th IEEE Real-Time Systems Symposium (RTSS WIP ), pp.,. [1] A. Tel, Introuction to Distribute Algorithms. Cambrige University Press,. [] H. Khalil, Nonlinear systems. Prentice Hall Upper Sale River, NJ,. [] [Online]. Available: pf files/ Wireless pf/--1 A MICAz.pf [] [Online]. Available:

Exponential Tracking Control of Nonlinear Systems with Actuator Nonlinearity

Exponential Tracking Control of Nonlinear Systems with Actuator Nonlinearity Preprints of the 9th Worl Congress The International Feeration of Automatic Control Cape Town, South Africa. August -9, Exponential Tracking Control of Nonlinear Systems with Actuator Nonlinearity Zhengqiang

More information

An Introduction to Event-triggered and Self-triggered Control

An Introduction to Event-triggered and Self-triggered Control An Introuction to Event-triggere an Self-triggere Control W.P.M.H. Heemels K.H. Johansson P. Tabuaa Abstract Recent evelopments in computer an communication technologies have le to a new type of large-scale

More information

BEYOND THE CONSTRUCTION OF OPTIMAL SWITCHING SURFACES FOR AUTONOMOUS HYBRID SYSTEMS. Mauro Boccadoro Magnus Egerstedt Paolo Valigi Yorai Wardi

BEYOND THE CONSTRUCTION OF OPTIMAL SWITCHING SURFACES FOR AUTONOMOUS HYBRID SYSTEMS. Mauro Boccadoro Magnus Egerstedt Paolo Valigi Yorai Wardi BEYOND THE CONSTRUCTION OF OPTIMAL SWITCHING SURFACES FOR AUTONOMOUS HYBRID SYSTEMS Mauro Boccaoro Magnus Egerstet Paolo Valigi Yorai Wari {boccaoro,valigi}@iei.unipg.it Dipartimento i Ingegneria Elettronica

More information

Lyapunov Functions. V. J. Venkataramanan and Xiaojun Lin. Center for Wireless Systems and Applications. School of Electrical and Computer Engineering,

Lyapunov Functions. V. J. Venkataramanan and Xiaojun Lin. Center for Wireless Systems and Applications. School of Electrical and Computer Engineering, On the Queue-Overflow Probability of Wireless Systems : A New Approach Combining Large Deviations with Lyapunov Functions V. J. Venkataramanan an Xiaojun Lin Center for Wireless Systems an Applications

More information

Throughput Optimal Control of Cooperative Relay Networks

Throughput Optimal Control of Cooperative Relay Networks hroughput Optimal Control of Cooperative Relay Networks Emun Yeh Dept. of Electrical Engineering Yale University New Haven, C 06520, USA E-mail: emun.yeh@yale.eu Ranall Berry Dept. of Electrical an Computer

More information

Generalizing Kronecker Graphs in order to Model Searchable Networks

Generalizing Kronecker Graphs in order to Model Searchable Networks Generalizing Kronecker Graphs in orer to Moel Searchable Networks Elizabeth Boine, Babak Hassibi, Aam Wierman California Institute of Technology Pasaena, CA 925 Email: {eaboine, hassibi, aamw}@caltecheu

More information

Least-Squares Regression on Sparse Spaces

Least-Squares Regression on Sparse Spaces Least-Squares Regression on Sparse Spaces Yuri Grinberg, Mahi Milani Far, Joelle Pineau School of Computer Science McGill University Montreal, Canaa {ygrinb,mmilan1,jpineau}@cs.mcgill.ca 1 Introuction

More information

6 General properties of an autonomous system of two first order ODE

6 General properties of an autonomous system of two first order ODE 6 General properties of an autonomous system of two first orer ODE Here we embark on stuying the autonomous system of two first orer ifferential equations of the form ẋ 1 = f 1 (, x 2 ), ẋ 2 = f 2 (, x

More information

Adaptive Gain-Scheduled H Control of Linear Parameter-Varying Systems with Time-Delayed Elements

Adaptive Gain-Scheduled H Control of Linear Parameter-Varying Systems with Time-Delayed Elements Aaptive Gain-Scheule H Control of Linear Parameter-Varying Systems with ime-delaye Elements Yoshihiko Miyasato he Institute of Statistical Mathematics 4-6-7 Minami-Azabu, Minato-ku, okyo 6-8569, Japan

More information

Time-of-Arrival Estimation in Non-Line-Of-Sight Environments

Time-of-Arrival Estimation in Non-Line-Of-Sight Environments 2 Conference on Information Sciences an Systems, The Johns Hopkins University, March 2, 2 Time-of-Arrival Estimation in Non-Line-Of-Sight Environments Sinan Gezici, Hisashi Kobayashi an H. Vincent Poor

More information

On Characterizing the Delay-Performance of Wireless Scheduling Algorithms

On Characterizing the Delay-Performance of Wireless Scheduling Algorithms On Characterizing the Delay-Performance of Wireless Scheuling Algorithms Xiaojun Lin Center for Wireless Systems an Applications School of Electrical an Computer Engineering, Purue University West Lafayette,

More information

'HVLJQ &RQVLGHUDWLRQ LQ 0DWHULDO 6HOHFWLRQ 'HVLJQ 6HQVLWLYLW\,1752'8&7,21

'HVLJQ &RQVLGHUDWLRQ LQ 0DWHULDO 6HOHFWLRQ 'HVLJQ 6HQVLWLYLW\,1752'8&7,21 Large amping in a structural material may be either esirable or unesirable, epening on the engineering application at han. For example, amping is a esirable property to the esigner concerne with limiting

More information

NOTES ON EULER-BOOLE SUMMATION (1) f (l 1) (n) f (l 1) (m) + ( 1)k 1 k! B k (y) f (k) (y) dy,

NOTES ON EULER-BOOLE SUMMATION (1) f (l 1) (n) f (l 1) (m) + ( 1)k 1 k! B k (y) f (k) (y) dy, NOTES ON EULER-BOOLE SUMMATION JONATHAN M BORWEIN, NEIL J CALKIN, AND DANTE MANNA Abstract We stuy a connection between Euler-MacLaurin Summation an Boole Summation suggeste in an AMM note from 196, which

More information

Closed and Open Loop Optimal Control of Buffer and Energy of a Wireless Device

Closed and Open Loop Optimal Control of Buffer and Energy of a Wireless Device Close an Open Loop Optimal Control of Buffer an Energy of a Wireless Device V. S. Borkar School of Technology an Computer Science TIFR, umbai, Inia. borkar@tifr.res.in A. A. Kherani B. J. Prabhu INRIA

More information

A Novel Integral-Based Event Triggering Control for Linear Time-Invariant Systems

A Novel Integral-Based Event Triggering Control for Linear Time-Invariant Systems 53rd IEEE Conference on Decision and Control December 15-17, 2014. Los Angeles, California, USA A Novel Integral-Based Event Triggering Control for Linear Time-Invariant Systems Seyed Hossein Mousavi 1,

More information

Switching Time Optimization in Discretized Hybrid Dynamical Systems

Switching Time Optimization in Discretized Hybrid Dynamical Systems Switching Time Optimization in Discretize Hybri Dynamical Systems Kathrin Flaßkamp, To Murphey, an Sina Ober-Blöbaum Abstract Switching time optimization (STO) arises in systems that have a finite set

More information

VIRTUAL STRUCTURE BASED SPACECRAFT FORMATION CONTROL WITH FORMATION FEEDBACK

VIRTUAL STRUCTURE BASED SPACECRAFT FORMATION CONTROL WITH FORMATION FEEDBACK AIAA Guiance, Navigation, an Control Conference an Exhibit 5-8 August, Monterey, California AIAA -9 VIRTUAL STRUCTURE BASED SPACECRAT ORMATION CONTROL WITH ORMATION EEDBACK Wei Ren Ranal W. Bear Department

More information

Computing Exact Confidence Coefficients of Simultaneous Confidence Intervals for Multinomial Proportions and their Functions

Computing Exact Confidence Coefficients of Simultaneous Confidence Intervals for Multinomial Proportions and their Functions Working Paper 2013:5 Department of Statistics Computing Exact Confience Coefficients of Simultaneous Confience Intervals for Multinomial Proportions an their Functions Shaobo Jin Working Paper 2013:5

More information

Nested Saturation with Guaranteed Real Poles 1

Nested Saturation with Guaranteed Real Poles 1 Neste Saturation with Guarantee Real Poles Eric N Johnson 2 an Suresh K Kannan 3 School of Aerospace Engineering Georgia Institute of Technology, Atlanta, GA 3332 Abstract The global stabilization of asymptotically

More information

Experimental Robustness Study of a Second-Order Sliding Mode Controller

Experimental Robustness Study of a Second-Order Sliding Mode Controller Experimental Robustness Stuy of a Secon-Orer Sliing Moe Controller Anré Blom, Bram e Jager Einhoven University of Technology Department of Mechanical Engineering P.O. Box 513, 5600 MB Einhoven, The Netherlans

More information

Chapter 6: Energy-Momentum Tensors

Chapter 6: Energy-Momentum Tensors 49 Chapter 6: Energy-Momentum Tensors This chapter outlines the general theory of energy an momentum conservation in terms of energy-momentum tensors, then applies these ieas to the case of Bohm's moel.

More information

Linear First-Order Equations

Linear First-Order Equations 5 Linear First-Orer Equations Linear first-orer ifferential equations make up another important class of ifferential equations that commonly arise in applications an are relatively easy to solve (in theory)

More information

Delay Limited Capacity of Ad hoc Networks: Asymptotically Optimal Transmission and Relaying Strategy

Delay Limited Capacity of Ad hoc Networks: Asymptotically Optimal Transmission and Relaying Strategy Delay Limite Capacity of A hoc Networks: Asymptotically Optimal Transmission an Relaying Strategy Eugene Perevalov Lehigh University Bethlehem, PA 85 Email: eup2@lehigh.eu Rick Blum Lehigh University Bethlehem,

More information

Survey Sampling. 1 Design-based Inference. Kosuke Imai Department of Politics, Princeton University. February 19, 2013

Survey Sampling. 1 Design-based Inference. Kosuke Imai Department of Politics, Princeton University. February 19, 2013 Survey Sampling Kosuke Imai Department of Politics, Princeton University February 19, 2013 Survey sampling is one of the most commonly use ata collection methos for social scientists. We begin by escribing

More information

Power Generation and Distribution via Distributed Coordination Control

Power Generation and Distribution via Distributed Coordination Control Power Generation an Distribution via Distribute Coorination Control Byeong-Yeon Kim, Kwang-Kyo Oh, an Hyo-Sung Ahn arxiv:407.4870v [math.oc] 8 Jul 204 Abstract This paper presents power coorination, power

More information

Approximate reduction of dynamic systems

Approximate reduction of dynamic systems Systems & Control Letters 57 2008 538 545 www.elsevier.com/locate/sysconle Approximate reuction of ynamic systems Paulo Tabuaa a,, Aaron D. Ames b, Agung Julius c, George J. Pappas c a Department of Electrical

More information

Laplacian Cooperative Attitude Control of Multiple Rigid Bodies

Laplacian Cooperative Attitude Control of Multiple Rigid Bodies Laplacian Cooperative Attitue Control of Multiple Rigi Boies Dimos V. Dimarogonas, Panagiotis Tsiotras an Kostas J. Kyriakopoulos Abstract Motivate by the fact that linear controllers can stabilize the

More information

Sliding mode approach to congestion control in connection-oriented communication networks

Sliding mode approach to congestion control in connection-oriented communication networks JOURNAL OF APPLIED COMPUTER SCIENCE Vol. xx. No xx (200x), pp. xx-xx Sliing moe approach to congestion control in connection-oriente communication networks Anrzej Bartoszewicz, Justyna Żuk Technical University

More information

arxiv: v1 [physics.flu-dyn] 8 May 2014

arxiv: v1 [physics.flu-dyn] 8 May 2014 Energetics of a flui uner the Boussinesq approximation arxiv:1405.1921v1 [physics.flu-yn] 8 May 2014 Kiyoshi Maruyama Department of Earth an Ocean Sciences, National Defense Acaemy, Yokosuka, Kanagawa

More information

State observers and recursive filters in classical feedback control theory

State observers and recursive filters in classical feedback control theory State observers an recursive filters in classical feeback control theory State-feeback control example: secon-orer system Consier the riven secon-orer system q q q u x q x q x x x x Here u coul represent

More information

Node Density and Delay in Large-Scale Wireless Networks with Unreliable Links

Node Density and Delay in Large-Scale Wireless Networks with Unreliable Links Noe Density an Delay in Large-Scale Wireless Networks with Unreliable Links Shizhen Zhao, Xinbing Wang Department of Electronic Engineering Shanghai Jiao Tong University, China Email: {shizhenzhao,xwang}@sjtu.eu.cn

More information

Separation Principle for a Class of Nonlinear Feedback Systems Augmented with Observers

Separation Principle for a Class of Nonlinear Feedback Systems Augmented with Observers Proceeings of the 17th Worl Congress The International Feeration of Automatic Control Separation Principle for a Class of Nonlinear Feeback Systems Augmente with Observers A. Shiriaev, R. Johansson A.

More information

APPPHYS 217 Thursday 8 April 2010

APPPHYS 217 Thursday 8 April 2010 APPPHYS 7 Thursay 8 April A&M example 6: The ouble integrator Consier the motion of a point particle in D with the applie force as a control input This is simply Newton s equation F ma with F u : t q q

More information

Approximate Reduction of Dynamical Systems

Approximate Reduction of Dynamical Systems Proceeings of the 4th IEEE Conference on Decision & Control Manchester Gran Hyatt Hotel San Diego, CA, USA, December 3-, 6 FrIP.7 Approximate Reuction of Dynamical Systems Paulo Tabuaa, Aaron D. Ames,

More information

Total Energy Shaping of a Class of Underactuated Port-Hamiltonian Systems using a New Set of Closed-Loop Potential Shape Variables*

Total Energy Shaping of a Class of Underactuated Port-Hamiltonian Systems using a New Set of Closed-Loop Potential Shape Variables* 51st IEEE Conference on Decision an Control December 1-13 212. Maui Hawaii USA Total Energy Shaping of a Class of Uneractuate Port-Hamiltonian Systems using a New Set of Close-Loop Potential Shape Variables*

More information

IPA Derivatives for Make-to-Stock Production-Inventory Systems With Backorders Under the (R,r) Policy

IPA Derivatives for Make-to-Stock Production-Inventory Systems With Backorders Under the (R,r) Policy IPA Derivatives for Make-to-Stock Prouction-Inventory Systems With Backorers Uner the (Rr) Policy Yihong Fan a Benamin Melame b Yao Zhao c Yorai Wari Abstract This paper aresses Infinitesimal Perturbation

More information

From Local to Global Control

From Local to Global Control Proceeings of the 47th IEEE Conference on Decision an Control Cancun, Mexico, Dec. 9-, 8 ThB. From Local to Global Control Stephen P. Banks, M. Tomás-Roríguez. Automatic Control Engineering Department,

More information

A Note on Exact Solutions to Linear Differential Equations by the Matrix Exponential

A Note on Exact Solutions to Linear Differential Equations by the Matrix Exponential Avances in Applie Mathematics an Mechanics Av. Appl. Math. Mech. Vol. 1 No. 4 pp. 573-580 DOI: 10.4208/aamm.09-m0946 August 2009 A Note on Exact Solutions to Linear Differential Equations by the Matrix

More information

The canonical controllers and regular interconnection

The canonical controllers and regular interconnection Systems & Control Letters ( www.elsevier.com/locate/sysconle The canonical controllers an regular interconnection A.A. Julius a,, J.C. Willems b, M.N. Belur c, H.L. Trentelman a Department of Applie Mathematics,

More information

A simplified macroscopic urban traffic network model for model-based predictive control

A simplified macroscopic urban traffic network model for model-based predictive control Delft University of Technology Delft Center for Systems an Control Technical report 9-28 A simplifie macroscopic urban traffic network moel for moel-base preictive control S. Lin, B. De Schutter, Y. Xi,

More information

Perfect Matchings in Õ(n1.5 ) Time in Regular Bipartite Graphs

Perfect Matchings in Õ(n1.5 ) Time in Regular Bipartite Graphs Perfect Matchings in Õ(n1.5 ) Time in Regular Bipartite Graphs Ashish Goel Michael Kapralov Sanjeev Khanna Abstract We consier the well-stuie problem of fining a perfect matching in -regular bipartite

More information

Optimal Variable-Structure Control Tracking of Spacecraft Maneuvers

Optimal Variable-Structure Control Tracking of Spacecraft Maneuvers Optimal Variable-Structure Control racking of Spacecraft Maneuvers John L. Crassiis 1 Srinivas R. Vaali F. Lanis Markley 3 Introuction In recent years, much effort has been evote to the close-loop esign

More information

19 Eigenvalues, Eigenvectors, Ordinary Differential Equations, and Control

19 Eigenvalues, Eigenvectors, Ordinary Differential Equations, and Control 19 Eigenvalues, Eigenvectors, Orinary Differential Equations, an Control This section introuces eigenvalues an eigenvectors of a matrix, an iscusses the role of the eigenvalues in etermining the behavior

More information

Multi-agent Systems Reaching Optimal Consensus with Time-varying Communication Graphs

Multi-agent Systems Reaching Optimal Consensus with Time-varying Communication Graphs Preprints of the 8th IFAC Worl Congress Multi-agent Systems Reaching Optimal Consensus with Time-varying Communication Graphs Guoong Shi ACCESS Linnaeus Centre, School of Electrical Engineering, Royal

More information

The total derivative. Chapter Lagrangian and Eulerian approaches

The total derivative. Chapter Lagrangian and Eulerian approaches Chapter 5 The total erivative 51 Lagrangian an Eulerian approaches The representation of a flui through scalar or vector fiels means that each physical quantity uner consieration is escribe as a function

More information

θ x = f ( x,t) could be written as

θ x = f ( x,t) could be written as 9. Higher orer PDEs as systems of first-orer PDEs. Hyperbolic systems. For PDEs, as for ODEs, we may reuce the orer by efining new epenent variables. For example, in the case of the wave equation, (1)

More information

Nonlinear Adaptive Ship Course Tracking Control Based on Backstepping and Nussbaum Gain

Nonlinear Adaptive Ship Course Tracking Control Based on Backstepping and Nussbaum Gain Nonlinear Aaptive Ship Course Tracking Control Base on Backstepping an Nussbaum Gain Jialu Du, Chen Guo Abstract A nonlinear aaptive controller combining aaptive Backstepping algorithm with Nussbaum gain

More information

A simple model for the small-strain behaviour of soils

A simple model for the small-strain behaviour of soils A simple moel for the small-strain behaviour of soils José Jorge Naer Department of Structural an Geotechnical ngineering, Polytechnic School, University of São Paulo 05508-900, São Paulo, Brazil, e-mail:

More information

Level Construction of Decision Trees in a Partition-based Framework for Classification

Level Construction of Decision Trees in a Partition-based Framework for Classification Level Construction of Decision Trees in a Partition-base Framework for Classification Y.Y. Yao, Y. Zhao an J.T. Yao Department of Computer Science, University of Regina Regina, Saskatchewan, Canaa S4S

More information

Connections Between Duality in Control Theory and

Connections Between Duality in Control Theory and Connections Between Duality in Control heory an Convex Optimization V. Balakrishnan 1 an L. Vanenberghe 2 Abstract Several important problems in control theory can be reformulate as convex optimization

More information

SYNCHRONOUS SEQUENTIAL CIRCUITS

SYNCHRONOUS SEQUENTIAL CIRCUITS CHAPTER SYNCHRONOUS SEUENTIAL CIRCUITS Registers an counters, two very common synchronous sequential circuits, are introuce in this chapter. Register is a igital circuit for storing information. Contents

More information

Lecture Introduction. 2 Examples of Measure Concentration. 3 The Johnson-Lindenstrauss Lemma. CS-621 Theory Gems November 28, 2012

Lecture Introduction. 2 Examples of Measure Concentration. 3 The Johnson-Lindenstrauss Lemma. CS-621 Theory Gems November 28, 2012 CS-6 Theory Gems November 8, 0 Lecture Lecturer: Alesaner Mąry Scribes: Alhussein Fawzi, Dorina Thanou Introuction Toay, we will briefly iscuss an important technique in probability theory measure concentration

More information

Topic 7: Convergence of Random Variables

Topic 7: Convergence of Random Variables Topic 7: Convergence of Ranom Variables Course 003, 2016 Page 0 The Inference Problem So far, our starting point has been a given probability space (S, F, P). We now look at how to generate information

More information

ELEC3114 Control Systems 1

ELEC3114 Control Systems 1 ELEC34 Control Systems Linear Systems - Moelling - Some Issues Session 2, 2007 Introuction Linear systems may be represente in a number of ifferent ways. Figure shows the relationship between various representations.

More information

Capacity Analysis of MIMO Systems with Unknown Channel State Information

Capacity Analysis of MIMO Systems with Unknown Channel State Information Capacity Analysis of MIMO Systems with Unknown Channel State Information Jun Zheng an Bhaskar D. Rao Dept. of Electrical an Computer Engineering University of California at San Diego e-mail: juzheng@ucs.eu,

More information

Event based Kalman filter observer for rotary high speed on/off valve

Event based Kalman filter observer for rotary high speed on/off valve 28 American Control Conference Westin Seattle Hotel, Seattle, Washington, USA June 11-13, 28 WeC9.6 Event base Kalman filter observer for rotary high spee on/off valve Meng Wang, Perry Y. Li ERC for Compact

More information

On the Connectivity Analysis over Large-Scale Hybrid Wireless Networks

On the Connectivity Analysis over Large-Scale Hybrid Wireless Networks This full text paper was peer reviewe at the irection of IEEE Communications Society subject matter experts for publication in the IEEE INFOCOM proceeings This paper was presente as part of the main Technical

More information

inflow outflow Part I. Regular tasks for MAE598/494 Task 1

inflow outflow Part I. Regular tasks for MAE598/494 Task 1 MAE 494/598, Fall 2016 Project #1 (Regular tasks = 20 points) Har copy of report is ue at the start of class on the ue ate. The rules on collaboration will be release separately. Please always follow the

More information

FORMATION INPUT-TO-STATE STABILITY. Herbert G. Tanner and George J. Pappas

FORMATION INPUT-TO-STATE STABILITY. Herbert G. Tanner and George J. Pappas Copyright 2002 IFAC 5th Triennial Worl Congress, Barcelona, Spain FORMATION INPUT-TO-STATE STABILITY Herbert G. Tanner an George J. Pappas Department of Electrical Engineering University of Pennsylvania

More information

APPROXIMATE SOLUTION FOR TRANSIENT HEAT TRANSFER IN STATIC TURBULENT HE II. B. Baudouy. CEA/Saclay, DSM/DAPNIA/STCM Gif-sur-Yvette Cedex, France

APPROXIMATE SOLUTION FOR TRANSIENT HEAT TRANSFER IN STATIC TURBULENT HE II. B. Baudouy. CEA/Saclay, DSM/DAPNIA/STCM Gif-sur-Yvette Cedex, France APPROXIMAE SOLUION FOR RANSIEN HEA RANSFER IN SAIC URBULEN HE II B. Bauouy CEA/Saclay, DSM/DAPNIA/SCM 91191 Gif-sur-Yvette Ceex, France ABSRAC Analytical solution in one imension of the heat iffusion equation

More information

Math Notes on differentials, the Chain Rule, gradients, directional derivative, and normal vectors

Math Notes on differentials, the Chain Rule, gradients, directional derivative, and normal vectors Math 18.02 Notes on ifferentials, the Chain Rule, graients, irectional erivative, an normal vectors Tangent plane an linear approximation We efine the partial erivatives of f( xy, ) as follows: f f( x+

More information

An Analytical Expression of the Probability of Error for Relaying with Decode-and-forward

An Analytical Expression of the Probability of Error for Relaying with Decode-and-forward An Analytical Expression of the Probability of Error for Relaying with Decoe-an-forwar Alexanre Graell i Amat an Ingmar Lan Department of Electronics, Institut TELECOM-TELECOM Bretagne, Brest, France Email:

More information

Optimal Cooperative Spectrum Sensing in Cognitive Sensor Networks

Optimal Cooperative Spectrum Sensing in Cognitive Sensor Networks Optimal Cooperative Spectrum Sensing in Cognitive Sensor Networks Hai Ngoc Pham, an Zhang, Paal E. Engelsta,,3, Tor Skeie,, Frank Eliassen, Department of Informatics, University of Oslo, Norway Simula

More information

05 The Continuum Limit and the Wave Equation

05 The Continuum Limit and the Wave Equation Utah State University DigitalCommons@USU Founations of Wave Phenomena Physics, Department of 1-1-2004 05 The Continuum Limit an the Wave Equation Charles G. Torre Department of Physics, Utah State University,

More information

PDE Notes, Lecture #11

PDE Notes, Lecture #11 PDE Notes, Lecture # from Professor Jalal Shatah s Lectures Febuary 9th, 2009 Sobolev Spaces Recall that for u L loc we can efine the weak erivative Du by Du, φ := udφ φ C0 If v L loc such that Du, φ =

More information

Optimal Control of Spatially Distributed Systems

Optimal Control of Spatially Distributed Systems Optimal Control of Spatially Distribute Systems Naer Motee an Ali Jababaie Abstract In this paper, we stuy the structural properties of optimal control of spatially istribute systems. Such systems consist

More information

Why Bernstein Polynomials Are Better: Fuzzy-Inspired Justification

Why Bernstein Polynomials Are Better: Fuzzy-Inspired Justification Why Bernstein Polynomials Are Better: Fuzzy-Inspire Justification Jaime Nava 1, Olga Kosheleva 2, an Vlaik Kreinovich 3 1,3 Department of Computer Science 2 Department of Teacher Eucation University of

More information

An M/G/1 Retrial Queue with Priority, Balking and Feedback Customers

An M/G/1 Retrial Queue with Priority, Balking and Feedback Customers Journal of Convergence Information Technology Volume 5 Number April 1 An M/G/1 Retrial Queue with Priority Balking an Feeback Customers Peishu Chen * 1 Yiuan Zhu 1 * 1 Faculty of Science Jiangsu University

More information

TRACKING CONTROL OF MULTIPLE MOBILE ROBOTS: A CASE STUDY OF INTER-ROBOT COLLISION-FREE PROBLEM

TRACKING CONTROL OF MULTIPLE MOBILE ROBOTS: A CASE STUDY OF INTER-ROBOT COLLISION-FREE PROBLEM 265 Asian Journal of Control, Vol. 4, No. 3, pp. 265-273, September 22 TRACKING CONTROL OF MULTIPLE MOBILE ROBOTS: A CASE STUDY OF INTER-ROBOT COLLISION-FREE PROBLEM Jurachart Jongusuk an Tsutomu Mita

More information

Minimum-time constrained velocity planning

Minimum-time constrained velocity planning 7th Meiterranean Conference on Control & Automation Makeonia Palace, Thessaloniki, Greece June 4-6, 9 Minimum-time constraine velocity planning Gabriele Lini, Luca Consolini, Aurelio Piazzi Università

More information

Harmonic Modelling of Thyristor Bridges using a Simplified Time Domain Method

Harmonic Modelling of Thyristor Bridges using a Simplified Time Domain Method 1 Harmonic Moelling of Thyristor Briges using a Simplifie Time Domain Metho P. W. Lehn, Senior Member IEEE, an G. Ebner Abstract The paper presents time omain methos for harmonic analysis of a 6-pulse

More information

Balancing Expected and Worst-Case Utility in Contracting Models with Asymmetric Information and Pooling

Balancing Expected and Worst-Case Utility in Contracting Models with Asymmetric Information and Pooling Balancing Expecte an Worst-Case Utility in Contracting Moels with Asymmetric Information an Pooling R.B.O. erkkamp & W. van en Heuvel & A.P.M. Wagelmans Econometric Institute Report EI2018-01 9th January

More information

Lecture 6: Calculus. In Song Kim. September 7, 2011

Lecture 6: Calculus. In Song Kim. September 7, 2011 Lecture 6: Calculus In Song Kim September 7, 20 Introuction to Differential Calculus In our previous lecture we came up with several ways to analyze functions. We saw previously that the slope of a linear

More information

Continuous observer design for nonlinear systems with sampled and delayed output measurements

Continuous observer design for nonlinear systems with sampled and delayed output measurements Preprints of th9th Worl Congress The International Feeration of Automatic Control Continuous observer esign for nonlinear systems with sample an elaye output measurements Daoyuan Zhang Yanjun Shen Xiaohua

More information

Perturbation Analysis and Optimization of Stochastic Flow Networks

Perturbation Analysis and Optimization of Stochastic Flow Networks IEEE TRANSACTIONS ON AUTOMATIC CONTROL, VOL. XX, NO. Y, MMM 2004 1 Perturbation Analysis an Optimization of Stochastic Flow Networks Gang Sun, Christos G. Cassanras, Yorai Wari, Christos G. Panayiotou,

More information

A PAC-Bayesian Approach to Spectrally-Normalized Margin Bounds for Neural Networks

A PAC-Bayesian Approach to Spectrally-Normalized Margin Bounds for Neural Networks A PAC-Bayesian Approach to Spectrally-Normalize Margin Bouns for Neural Networks Behnam Neyshabur, Srinah Bhojanapalli, Davi McAllester, Nathan Srebro Toyota Technological Institute at Chicago {bneyshabur,

More information

The Principle of Least Action

The Principle of Least Action Chapter 7. The Principle of Least Action 7.1 Force Methos vs. Energy Methos We have so far stuie two istinct ways of analyzing physics problems: force methos, basically consisting of the application of

More information

Equilibrium in Queues Under Unknown Service Times and Service Value

Equilibrium in Queues Under Unknown Service Times and Service Value University of Pennsylvania ScholarlyCommons Finance Papers Wharton Faculty Research 1-2014 Equilibrium in Queues Uner Unknown Service Times an Service Value Laurens Debo Senthil K. Veeraraghavan University

More information

Reactive Power Compensation in Mechanical Systems

Reactive Power Compensation in Mechanical Systems Reactive Power Compensation in Mechanical Systems Carlos Rengifo, Bassel Kaar, Yannick Aoustin, Christine Chevallereau To cite this version: Carlos Rengifo, Bassel Kaar, Yannick Aoustin, Christine Chevallereau.

More information

Event-Triggered Output Feedback Control for Networked Control Systems using Passivity: Time-varying Network Induced Delays

Event-Triggered Output Feedback Control for Networked Control Systems using Passivity: Time-varying Network Induced Delays 5th IEEE Conference on Decision and Control and European Control Conference (CDC-ECC) Orlando, FL, USA, December -5, Event-Triggered Output Feedback Control for Networked Control Systems using Passivity:

More information

Math 342 Partial Differential Equations «Viktor Grigoryan

Math 342 Partial Differential Equations «Viktor Grigoryan Math 342 Partial Differential Equations «Viktor Grigoryan 6 Wave equation: solution In this lecture we will solve the wave equation on the entire real line x R. This correspons to a string of infinite

More information

Quantum Search on the Spatial Grid

Quantum Search on the Spatial Grid Quantum Search on the Spatial Gri Matthew D. Falk MIT 2012, 550 Memorial Drive, Cambrige, MA 02139 (Date: December 11, 2012) This paper explores Quantum Search on the two imensional spatial gri. Recent

More information

A Simple Model for the Calculation of Plasma Impedance in Atmospheric Radio Frequency Discharges

A Simple Model for the Calculation of Plasma Impedance in Atmospheric Radio Frequency Discharges Plasma Science an Technology, Vol.16, No.1, Oct. 214 A Simple Moel for the Calculation of Plasma Impeance in Atmospheric Raio Frequency Discharges GE Lei ( ) an ZHANG Yuantao ( ) Shanong Provincial Key

More information

Concentration of Measure Inequalities for Compressive Toeplitz Matrices with Applications to Detection and System Identification

Concentration of Measure Inequalities for Compressive Toeplitz Matrices with Applications to Detection and System Identification Concentration of Measure Inequalities for Compressive Toeplitz Matrices with Applications to Detection an System Ientification Borhan M Sananaji, Tyrone L Vincent, an Michael B Wakin Abstract In this paper,

More information

Yaakov (Jonathan) Stein, RAD Data Communications and Brian Stroehlein, TranSwitch Corporation

Yaakov (Jonathan) Stein, RAD Data Communications and Brian Stroehlein, TranSwitch Corporation Using ynchronization over PN Does IEEE 1588 M Really Mae a Difference? Yaaov (Jonathan) tein, RAD Data Communications an Brian troehlein, ranwitch Corporation pecial thans to Alon Geva, DP Algorithm eam

More information

arxiv: v1 [math.co] 29 May 2009

arxiv: v1 [math.co] 29 May 2009 arxiv:0905.4913v1 [math.co] 29 May 2009 simple Havel-Hakimi type algorithm to realize graphical egree sequences of irecte graphs Péter L. Erős an István Miklós. Rényi Institute of Mathematics, Hungarian

More information

Implicit Lyapunov control of closed quantum systems

Implicit Lyapunov control of closed quantum systems Joint 48th IEEE Conference on Decision an Control an 28th Chinese Control Conference Shanghai, P.R. China, December 16-18, 29 ThAIn1.4 Implicit Lyapunov control of close quantum systems Shouwei Zhao, Hai

More information

KNN Particle Filters for Dynamic Hybrid Bayesian Networks

KNN Particle Filters for Dynamic Hybrid Bayesian Networks KNN Particle Filters for Dynamic Hybri Bayesian Networs H. D. Chen an K. C. Chang Dept. of Systems Engineering an Operations Research George Mason University MS 4A6, 4400 University Dr. Fairfax, VA 22030

More information

Stopping-Set Enumerator Approximations for Finite-Length Protograph LDPC Codes

Stopping-Set Enumerator Approximations for Finite-Length Protograph LDPC Codes Stopping-Set Enumerator Approximations for Finite-Length Protograph LDPC Coes Kaiann Fu an Achilleas Anastasopoulos Electrical Engineering an Computer Science Dept. University of Michigan, Ann Arbor, MI

More information

Calculus of Variations

Calculus of Variations 16.323 Lecture 5 Calculus of Variations Calculus of Variations Most books cover this material well, but Kirk Chapter 4 oes a particularly nice job. x(t) x* x*+ αδx (1) x*- αδx (1) αδx (1) αδx (1) t f t

More information

Time Headway Requirements for String Stability of Homogeneous Linear Unidirectionally Connected Systems

Time Headway Requirements for String Stability of Homogeneous Linear Unidirectionally Connected Systems Joint 48th IEEE Conference on Decision an Control an 8th Chinese Control Conference Shanghai, PR China, December 6-8, 009 WeBIn53 Time Heaway Requirements for String Stability of Homogeneous Linear Uniirectionally

More information

Hybrid Centralized-Distributed Resource Allocation for Device-to-Device Communication Underlaying Cellular Networks

Hybrid Centralized-Distributed Resource Allocation for Device-to-Device Communication Underlaying Cellular Networks Hybri Centralize-Distribute Resource Allocation for Device-to-Device Communication Unerlaying Cellular Networks 1 arxiv:1502.04539v1 [cs.gt] 16 Feb 2015 Setareh Maghsui an Sławomir Stańczak, Senior Member,

More information

Cascaded redundancy reduction

Cascaded redundancy reduction Network: Comput. Neural Syst. 9 (1998) 73 84. Printe in the UK PII: S0954-898X(98)88342-5 Cascae reunancy reuction Virginia R e Sa an Geoffrey E Hinton Department of Computer Science, University of Toronto,

More information

Comparative Approaches of Calculation of the Back Water Curves in a Trapezoidal Channel with Weak Slope

Comparative Approaches of Calculation of the Back Water Curves in a Trapezoidal Channel with Weak Slope Proceeings of the Worl Congress on Engineering Vol WCE, July 6-8,, Lonon, U.K. Comparative Approaches of Calculation of the Back Water Curves in a Trapezoial Channel with Weak Slope Fourar Ali, Chiremsel

More information

Systems & Control Letters

Systems & Control Letters Systems & ontrol Letters ( ) ontents lists available at ScienceDirect Systems & ontrol Letters journal homepage: www.elsevier.com/locate/sysconle A converse to the eterministic separation principle Jochen

More information

Optimal Control of Spatially Distributed Systems

Optimal Control of Spatially Distributed Systems Optimal Control of Spatially Distribute Systems Naer Motee an Ali Jababaie Abstract In this paper, we stuy the structural properties of optimal control of spatially istribute systems. Such systems consist

More information

Optimal Signal Detection for False Track Discrimination

Optimal Signal Detection for False Track Discrimination Optimal Signal Detection for False Track Discrimination Thomas Hanselmann Darko Mušicki Dept. of Electrical an Electronic Eng. Dept. of Electrical an Electronic Eng. The University of Melbourne The University

More information

Fast Structuring of Radio Networks

Fast Structuring of Radio Networks Fast Structuring of Raio Networks for Multi-Message Communications Mohsen Ghaffari ghaffari@mit.eu MIT Bernhar Haeupler haeupler@mit.eu MIT Abstract We introuce collision free layerings as a powerful way

More information

The Optimal Steady-State Control Problem

The Optimal Steady-State Control Problem SUBMITTED TO IEEE TRANSACTIONS ON AUTOMATIC CONTROL. THIS VERSION: OCTOBER 31, 218 1 The Optimal Steay-State Control Problem Liam S. P. Lawrence Stuent Member, IEEE, John W. Simpson-Porco, Member, IEEE,

More information

u!i = a T u = 0. Then S satisfies

u!i = a T u = 0. Then S satisfies Deterministic Conitions for Subspace Ientifiability from Incomplete Sampling Daniel L Pimentel-Alarcón, Nigel Boston, Robert D Nowak University of Wisconsin-Maison Abstract Consier an r-imensional subspace

More information