Managing non-determinism in symbolic robot motion planning and control
|
|
- Hilary Matthews
- 5 years ago
- Views:
Transcription
1 2007 IEEE Interntionl Conference on Robotics nd Automtion Rom, Itly, April 2007 ThD11.4 Mnging non-determinism in symbolic robot motion plnning nd control Mrius Kloetzer nd Clin Belt Center for Informtion nd Systems Engineering Boston University 15 Sint Mry s Street, Boston, MA {kmrius,cbelt}@bu.edu Abstrct We study the problem of designing control strtegies for non-deterministic trnsitions systems enforcing the stisfction of Liner Temporl Logic (LTL) formuls over their set of sttes. We focus on nite trnsition systems with inputs, which re often encountered when solving motion plnning problems by using discrete quotients induced by given prtition of the stte spce. Our pproch solves the problem conservtively using LTL gmes, nd consists of the following three steps: (1) the originl trnsition system is trnsformed into trnsition system on which n LTL gme cn be plyed, (2) solution of the LTL gme on the new trnsition system is obtined, nd (3) n interfce between this solution nd the initil trnsition system is constructed. The correctness of the method is ensured by design. The dvntges nd conservtiveness of our pproch re discussed nd illustrted by simple exmples. I. INTRODUCTION Motion plnning nd control of robots with nontrivil dynmics or kinemtics is usully two step process. In the rst step, (cellulr) decomposition of the C-spce is constructed, nd discrete pth is generted by serch in the quotient grph [1], [2]. In the second step, reference trjectory comptible with the robot dynmics or kinemtic is generted, nd robot control lws re designed to follow the trjectory. A very ttrctive lterntive to this is simultneous plnning nd control, in which the genertion of the discrete solution in the prtition quotient is performed t the sme time with the ssignment of vector elds in the regions of the prtition (robot control lws), while tking into ccount the restrictions imposed by robot under-ctution nd speed constrints [3], [4], [5], [6]. In ddition, enrichment of the speci ction lnguge from the clssicl plnning tsk go from A to B to temporl logic speci ctions (e.g., visit either A or B, rech A nd then B in nitely often ) leds to symbolic pproches to simultneous plnning nd control, where discrete bstrctions re used to provide forml link between the continuous robot control system nd the discrete representtion of the environment, nd lgorithms resembling model checking re used to provide solution to the discrete problem [7], [8]. All the pproches enumerted bove for simultneous plnning nd control fce common problem: the restrictions imposed by the robot dynmics or kinemtics cn, in generl, led to non-deterministic representtions of the discrete problem. For exmple, the prepres reltionship between collection of policies in [4] results in nondeterminism of the discrete bstrction. Recent results in control of f ne systems in simplices [9] nd of multi-f ne systems in rectngles [10] show tht even though controllers determining the trnsition of robot to speci c neighbor might fil to exist, controllers gurnteeing the trnsition to set of neighbors cn be found. In the trnsition system corresponding to the discrete prt of the problem, this mens non-determinism. Motivted by the bove, in this pper we focus on the discrete prt of simultneous motion plnning nd control, nd consider the following problem: given non-deterministic trnsition system with inputs, nd Liner Temporl Logic (LTL) formul over its set of sttes, determine set of initil sttes nd control strtegy so tht the produced trjectory stis es the formul. This problem is quite generl, nd, to the best of our knowledge, there is no vilble computtionl frmework providing solution. However, it is relted to LTL gmes [11]. Also, it is possible tht solutions cn be found by using results from the control of discrete event systems from speci ctions given s ω- regulr expressions (lnguges) over inputs [12] or by using tree utomt on in nite objects [13]. In this work, we dvocte the use of LTL gmes [11] for the construction of (conservtive) solution to the problem. Our fully utomtic computtionl frmework consists of three min steps. First, we convert the originl trnsition system into form in which n LTL gme cn be plyed. Second, we nd solution to the LTL gme in the form of feedbck utomton. Third, the solution of the gme is used to generte control strtegy for the initil trnsition system. We illustrte our pproch for the cse of tringulted plnr environment, where the ssignment of f ne feedbck controllers in the tringles using the method developed in [9] leds to non-determinism. The reminder of the pper is orgnized s follows. Section II provides some de nitions necessry throughout the pper. The problem is formulted in Section III nd our solution is presented in Section IV. Simultion results re given in Section V nd we conclude with nl remrks in Section VI /07/$ IEEE. 3110
2 II. PRELIMINARIES For nite set A, we use A, A ω, nd 2 A to denote its crdinlity, the set of ll in nite words over A, nd its power set (the set of ll its subsets), respectively. A. Trnsition Systems nd Liner Temporl Logic De nition 1: [Non-deterministic trnsition system] A - nite non-deterministic trnsition system is tuple T = (Q, Σ, δ, h, O), where: Q is nite set of sttes, Σ is nite input lphbet, δ : Q Σ 2 Q is trnsition function, h : Q O is the observtion mp, O is the set of observbles. For given stte q Q, the set of vilble (fesible) inputs is denoted by Σ q (i.e., Σ q is the set of σ i Σ for which δ(q, σ i ) 1). An input word σ Σ ω is denoted by σ = σ 1 σ 2 σ 3... A trjectory or run of T produced by n input word σ strting from q is n in nite sequence r Q ω, r = r 1 r 2 r 3... with the property tht r 1 = q nd i 1, r i+1 δ(r i,σ i ). A trjectory r of T produces word w O ω de ned s w = w 1 w 2 w 3..., w i = h(q i ), for ll i 1. A trnsition system T = (Q, Σ, δ, h, O) for which the observtion mp is identity (i.e., the sttes re of interest Q = O, nd cn be observed) is denoted for simplicity by T =(Q, Σ,δ). De nition 2: [Syntx of LTL formuls] An LTL formul over O is recursively de ned s follows: Every observble o O is formul, nd If φ 1 nd φ 2 re formuls, then φ 1, φ 1 φ 2, φ 1, φ 1 Uφ 2 re lso formuls. The semntics of LTL formuls re given over words of trnsition system T. De nition 3: [Semntics of LTL formuls] The stisfction of formul φ t position i N of word w, denoted by w i φ, is de ned recursively s follows: w i o if o = w i, w i φ if w i φ, w i φ 1 φ 2 if w i φ 1 or w i φ 2, w i φ if w i+1 φ, w i φ 1 Uφ 2 if there exist j i such tht w j φ 2 nd for ll i k < j we hve w k φ 1 A word w stis es n LTL formul φ, written s w φ, if w 1 φ. The symbols nd stnd for negtion nd disjunction. The Boolen constnts nd re de ned s = π π nd =. The other Boolen connectors (conjunction), (impliction), nd (equivlence) re de ned from nd in the usul wy. The unry temporl opertor is clled next, nd the binry temporl opertor U is clled the until opertor. Two useful dditionl temporl opertors, eventully nd lwys cn be de ned s φ = Uφ nd φ = φu, respectively. Formul φ mens tht φ becomes eventully true, wheres φ indictes tht φ is true t ll positions of trjectory. Remrk 1: In generl, the semntics of LTL formuls re given over in nite words in the power set of set of tomic propositions. We use the simpli ed semntics of De nition 3 motivted by the problem formulted in Section III. ABüchi utomton is nite stte utomton ccepting in nite strings. The de nition of Büchi utomton nd its cceptnce condition is beyond the scope of this pper, nd we refer the interested reder to [13]. For ny LTL formul, there exists non-deterministic Büchi utomton ccepting ll nd only the runs stisfying it [14] (this utomton is lso clled genertor of the LTL formul). B. LTL Gmes An LTL gme is de ned on grph G =(V, E, h, O), where V is the set of vertices, E V V is the set of edges, O is set of observbles, nd h : V O is n observtion mp. The gme speci ction is n LTL formul over the observble set O [15] (see lso [11] for the cse of grphs with no observbles). There re two plyers in the gme: P (the plyer) nd A (the dversry). The set of vertices V is prtitioned in two sets: V p, which is the set of sttes from which the plyer cn choose move (n edge leding to the next vertex), nd V, which is the set of sttes from which the dversry hs choice of n edge. We ssume tht the current vertex from set V is lwys known, nd not just its observble. An (in nite) ply consists of n in nite sequence of sttes resulted from n in nite sequence of trnsitions (edges) chosen by the two plyers. The plyer P wins ply if the produced word w O ω stis es the LTL formul. Strting from given initil stte, the plyer hs winning strtegy if, whenever the current stte is in V p, she mnges to choose edges such tht she wins the current ply, no mtter wht edges the dversry chooses when the current stte is in V. The gol of n LTL gme is to nd set of initil sttes W s V from where the plyer hs winning strtegies nd winning strtegy for plys strting in those initil sttes. The stndrd lgorithm for solving n LTL gme involves the trnsformtion of the non-deterministic Büchi utomton corresponding to the LTL formul into deterministic genertor [16]. Motivted by the simplicity of exposition nd by some complexity issues, we ssume tht the LTL formuls we re deling with ccept deterministic Büchi genertors (there re cses when this is not true [17], nd lgorithms for solving LTL gmes become more complicted). For more detils of the steps involved in solving LTL gmes the interested reder is referred to [16], [15]. The result of solving n LTL gme on G =(V, E, h, O) (s outlined bove) is set of initil sttes W s V (if non-empty) nd winning strtegy W, which is strtegy (s in De nition 4) gurnteeing the stisfction of the LTL formul whenever the initil stte is in the set W s. De nition 4: A strtegy is n utomton W = (S, V, O, s 0, τ, π), where: S is the nite set of sttes (the memory), V is the input lphbet, O is the set over which the LTL formul φ ws de ned, 3111
3 s 0 S is the initil stte (initil memory), τ : S O S is the memory updte function, π : S V p V is the function giving the chosen move when the current stte of G is in V p. The sttes S result from the sttes of the deterministic utomton (e.g., Büchi) corresponding to the LTL formul. The current vertex of G gives the input of W nd it is used for generting the plyer s moves, while the corresponding observble is used for updting the internl memory of W. Thus, given the current stte (memory) of W, the strtegy π depends on the current stte of G, while the memory updte function τ depends only on the observble of this stte. The reltion between grphs nd trnsition systems is immedite. When plying n LTL gme on trnsition system, insted of choosing the next stte, we choose n input producing the desired trnsition. This is possible if plyer s sttes hve only deterministic outgoing trnsitions, nd then the function π from the winning strtegy will tke vlues in the input set of the trnsition system insted of V. III. PROBLEM FORMULATION AND APPROACH In this pper we consider the following problem: Problem 1: Given non-deterministic trnsition system T =(Q, Σ,δ) nd n LTL formul φ over Q, nd set of fesible initil sttes Q 0 Q nd control strtegy such tht formul φ is stis ed by ll resulting in nite words. Remrk 2: If the trnsition system T ws deterministic (i.e., δ : Q Σ Q), then solution to Problem 1 could be found by using n ide similr to model checking [18], [7], [19]: the LTL formul φ is trnsformed into Büchi utomton B φ nd the product utomton T B φ is computed. In this product utomton n ccepting run with speci c structure is found nd projected to run of T. Since T is deterministic, control strtegy implementing the desired run cn be constructed. We propose to use the frmework of LTL gmes to provide solution to Problem 1. For this, we need to reformulte the problem from grph to trnsition system, nd then de ne prtition of the set of sttes into plyer s sttes nd dversry s sttes. Remrk 3: One quick wy of solving this would be to lbel s dversry s stte ech stte from where there exists t lest one input yielding non-determinist trnsition. However, such n pproch would be pretty conservtive, becuse if such n dversry s stte hd more thn one fesible input, we wouldn t use our freedom of choosing one from these inputs nd thus restricting the set of possible next sttes. Our pproch involves three min steps. First, we crete new trnsition system T g with observbles, where we choose the plyer s sttes by looking for sttes nd inputs with deterministic behvior. T g will, in generl, hve more sttes nd inputs thn T, nd its observble set will be Q. Second, we serch for winning strtegy for the LTL gme on T g. Finlly, when such strtegy exists, we dpt it to control strtegy providing t ech step the input to be pplied to our initil trnsition system T. IV. CONTROL STRATEGY USING LTL GAMES We strt by trnsforming T into nother trnsition system T g =(Q g, Σ g,δ g,h g,q), where: Q g is the nite set of sttes, prtitioned into sets Q p (plyer s sttes) nd Q (dversry s sttes), Σ g is the new lphbet, Σ Σ g, δ g : Q g Σ g 2 Qg is the trnsition function, h g : Q g Q is the observtion mp. The construction of T g is described in Section IV-A. Here we give some de nitions nd outline how the gme is plyed. The trnsition function δ g is de ned s: its restriction to Q p Σ g is deterministic, i.e. q Q p nd σ Σ g, δ g (q, σ) 1, its restriction to Q Σ g is non-deterministic, nd there is only one fesible (enbled) input in every stte from Q, i.e. q Q, there exists n unique input σ Σ g such tht δ g (q, σ) 2 nd ll other inputs σ Σ g \ {σ} re not fesible t q (or δ g (q, σ ) =0). These properties of δ g show tht when the current stte is in Q p, the plyer P cn lwys choose the next stte from fesible set (by pplying fesible input), nd when the current stte is in Q, P hs no control on the next stte (we only know tht the next stte will be in subset of Q g, given by the unique fesible input). Thus, the structure of T g is pproprite for n LTL gme s described in Section II-B. By solving the LTL gme over T g with winning condition φ, one cn obtin set of winning initil sttes, W s Q g, nd, if W s, winning strtegy in the form of nite utomton, similr to the one from De nition 4. When W s is nonempty, the winning strtegy is W =(S, Q g, Q, s 0, τ, π), where: S is nite set of sttes, Q g is the input lphbet, Q is the set over which the LTL formul φ ws de ned, s 0 S is the initil stte, τ : S Q S is the memory updte function, π : S Q p Σ g is the function giving the choice of the next pplied input (move) when the current stte of T g is in the set Q p. The difference between this winning strtegy nd the one from De nition 4 is tht here π tkes vlues in Σ g insted of Q g. Becuse of the mentioned properties of δ g, every move the plyer should mke when the ply is in Q p corresponds to n unique input from Σ g. In order to obtin n input to be pplied to T g in every stte, the mp π is extended to S Q g, by letting π(s, q) =σ, where σ Σ g is the only fesible input in stte q, q Q. The winning strtegy given by utomton W is implemented in feedbck form. At ech step, the following ctions re performed ( step is the intervl between two successive trnsitions of T g ): (1) it reds the current stte of T g, (2) ccording to function π, it pplies n input from the set Σ g to T g, nd (3) it updtes its stte ccording to function τ. If the initil stte of T g is in the set W s, this feedbck control strtegy gurntees tht the gme will be won. Note 3112
4 q 2 q 3 q 4 b () 1 1 q 2 b b 2 q 4 3 q 3 b q 4 1 q 4 Fig. 1. () Trnsition system T : nd q 3 re det sttes, q 2 is ndet, nd q 4 is mix. (b) Obtined T g: q 4 from T ws split in two ndet sttes in T g nd two new inputs ppered in Σ g. In T g, the observbles re plced ner the sttes, nd the dversry s sttes re doubly encircled. All deterministic trnsitions re represented with continuous lines, non-deterministic ones with dshed lines, nd ech input is plced ner the corresponding trnsition(s). tht this feedbck controller ssumes tht we cn lwys red the current stte of T g nd not only its observble. A. Construction of T g We present here the min ides used for trnsforming the initil trnsition system T into the new trnsition system T g. Due to spce constrints, we do not include ny lgorithm nd we refer to [20] for full pseudo-codes. For creting the sttes nd the observtion mp of T g, we lbel the sttes from Q by function l : Q {det, ndet, mix}: stte with only deterministic outgoing trnsitions or without outgoing trnsitions is lbelled with det, stte with only one fesible input, for which the outgoing trnsitions re non-deterministic, is lbelled with ndet, nd stte with more fesible inputs, from which t lest one yields non-deterministic trnsitions, is lbelled with mix. Ech mix stte is split in one or more ndet sttes nd t most one det stte in the new trnsition system T g (for simplicity of writing, we ssume tht the ttributes ndet nd det for sttes of T g hve exctly the sme mening s in T ). The observble of ll these new sttes is the old mix stte. Ech det (respectively ndet) stte from T corresponds to det (respectively ndet) stte in T g. Thus, T g will hve only det (in the subset Q p ) nd ndet sttes (in the subset Q ). For better understnding we present in Fig. 1 simple exmple of trnsforming trnsition system T into T g. The input lphbet Σ g nd the trnsition function δ g re creted fter the set of sttes Q g = Q p Q is obtined. T g hs lrger input set thn T becuse the det sttes from T must remin det in T g. This cn be explined by following the exmple in Fig. 1: in T, the det stte q 3 hs trnsition with input b to the mix stte q 4. q 3 corresponds to one stte of T g ( 3), but q 4 is split in two sttes ( 4 nd q 2 4). Since we need deterministic trnsitions for going from 3 to either 4 or q 2 4, we introduce the new inputs b 1 nd b 2. Thus, by pplying b 1 when T g is in 3, the next stte is 4 nd there is only one next fesible input (). In T this hs the effect of pplying input b while in q 3 nd enforcing input when q 4 is reched. Similrly, when b 2 is pplied to T g while in 3, this induces the sequence of inputs b, b strting from q 3 in T. Although the dversry hs full control on the trnsitions from 4 nd q 2 4 in T g, the plyer hs control in going to either b 2 b 1 (b) q 4 q 2 4 or q 2 4, thus restricting the dversry s power. Following the sme ide, ech det stte of T g (including those obtined by splitting mix sttes of T ) might dd some new inputs to the lphbet Σ g. In order to dpt these new inputs to inputs of T, mp α :Σ g Σ is creted, which will be used by the controller for T. For ll inputs in set Σ, α is just the identity mp. The ndet sttes of T g (from dversry s set Q ) will not dd new inputs. Ech of these sttes will hve only one fesible input (from set Σ), nd thus the dversry power in these sttes comes only from the non-determinism induced by tht fesible input. B. Solution to Problem 1 We now hve ll the necessry informtion to present the solution to Problem 1. The set of initil sttes of T from where the controller gurntees the stisfction of the LTL formul φ is: Q 0 = {q Q q W s s.t.h g (q )=q} (1) The set Q 0 from (1) is in ccordnce with the solution of the LTL gme for the trnsition system T g : if the initil stte of T g is in the set W s, then the strtegy generted by utomton W gurntees the stisfction of formul φ by ny produced run of T g. Of course, Q 0 = if nd only if W s =, cse when we conclude tht we cnnot solve Problem 1. If Q 0, the control strtegy for the trnsition system T will be feedbck controller, which will lso include the interfce between T nd the ply on T g, supervised by W. Agin, we brie y describe the opertions performed by this controller nd we refer to [20] for n lgorithm ccompnied by detiled explntions. At ech discrete step, correct stte of T g is rst picked: this stte must be in concordnce with the previous stte of T g nd the previously pplied input, nd its observble must equl the current stte of T. Then, bsed on the winning strtegy W, n input for T g is found, which is then dpted to n input for T by using mp α. Whenever the resulted input for T g is in the set Σ g \Σ, the dversry s power in the next stte of T will be reduced (for exmple, see the cse depicted in Fig. 1, when input b 1 or b 2 ppers). For better understnding, we complete the exmple from Fig. 1 by considering the formul φ = (q 4 ), mening visit q 4 nd then. By using the presented pproch, we obtin Q 0 = {q 3,q 4 }. When T strts from q 3, the rst two inputs pplied by the controller re b, (corresponding to input b 1 of T g ). If the resulted stte is q 3, the input will be pplied nd is reched. After is visited, the controller will keep pplying ny of the fesible inputs in the visited sttes (in our implementtion, will be lwys pplied). Note tht by trying to solve the sme problem by using the pproch from either Remrk 2 or Remrk 3, we would get no solution for the problem (in the rst cse there re no deterministic trnsitions out of q 4, nd in the second cse, once q 4 is reched, the dversry cn keep choosing the self-loop in this stte). 3113
5 C. Conservtiveness nd Complexity Our pproch is conservtive in the following sense: when creting the trnsitions of T g from ndet (dversry s) sttes, we don t restrict the set of inputs to be pplied to T t the next discrete step. For illustrting this, consider the exmple in Fig. 1 nd the formul φ =. Our pproch gives Q 0 = {,q 3,q 4 }, while one cn observe tht φ cn be lso stis ed by strting from q 2 nd keep pplying input, which eventully leds to being visited. However, pproches from Remrks 2 nd 3 give only strtegies from Q 0 = {,q 3 }. This problem does not pper in the cse of det sttes of T g, becuse of the inputs in {Σ g \ Σ}, which restricts the set of choices for the next input. If there re no trnsitions between mix nd ndet sttes of T, then this source of conservtiveness does not pper. Another source of conservtiveness in our pproch is tht we ssumed n LTL formul over the sttes of T.A more generl setting would involve trnsition system T with n observtion mp nd n LTL formul over the set of observbles. If sttes of T hving the sme observble were indistinguishble (not like in the cse of T g ), the proposed method wouldn t be suitble, becuse we couldn t nd stte of T g (with the correct observble) nd be sure tht the fesible inputs in tht stte re lso fesible in the rel stte of T. However, this formultion is motivted by our prticulr motion plnning ppliction, where we ssume tht the environment is prtitioned nd the resulting regions re lbelled (see exmple in Section V). Finlly, when our pproch fils in providing solution, we declre Problem 1 to be unfesible. Other pproches could use n itertive procedure obtining t ech step ner discrete representtion (tilored to speci c problem) nd trying to solve Problem 1 for tht re ned trnsition system. The upper bound complexity of the presented pproch is in generl high, becuse n LTL gme on grph T g is solvble in polynomil time in the size of the grph nd in double-exponentil time in the size of the LTL formul [21], [22]. However, there hve been isolted some LTL frgments ccepting deterministic Büchi utomt nd gurnteeing much lower complexity [22]. V. CASE STUDY To illustrte the pproch proposed in the pper, we consider simple plnning nd control problem for plnr continuous system evolving in rectngulr tringulted environment (see Fig. 2). The dynmics, rectngulr bounds, nd control constrints re given in eqution (2). We ssume tht the environment is prtitioned in 8 tringles (lbelled by q i,i=1,...,8 in Fig. 2), nd the tsk is speci ed by the LTL formul φ = q 4 q 7. In other words, the tsk requires tht regions q 4 nd q 7 be visited (in ny order) by the trjectories of the closed loop system. [ ] [ ] [ ẋ = x + u x [ 3, 4] [ 3, 3], u [ 1, 1] [ 1, 1] ], (2) !1!2 q 4 q 2 q 3!3!3!2! Fig. 2. Tringulr prtition (thin blck lines) nd two continuous trjectories strting from region q 3 nd stisfying formul φ. 4 q q 3 q 2 17 q Fig. 3. Trnsition system T corresponding to the prtition in Fig. 2. Deterministic trnsitions re represented with continuous lines, nondeterministic ones with dshed lines, nd ech input is plced ner the corresponding trnsition(s). A forml de nition of the semntics of the formul over continuous trjectories of the closed loop system is beyond the scope of this pper, nd cn be found in [19]. However, intuitively, trjectory stis es the formul if the word produced by the sequence of regions visited by the system s time evolves stis es the formul. As in [7], [19], we strt by constructing trnsition system T with sttes Q = {,...,q 8 } corresponding to the prtition elements nd with trnsitions cpturing the bility of designing f ne feedbck controllers such tht tringle is either left in nite time to neighbor, or becomes n invrint for the closed loop system. The construction of such controllers is computtionlly ef cient, since it cn be reduced to opertions on polyhedrl sets [19]. However, unlike in [7], [19], we don t restrict our ttention to controllers mking tringle n invrint or driving ll initil sttes in tringle to one neighbor (this utomticlly induces deterministic quotient, with trnsitions lbelled by the corresponding controllers). Enbled by recent results on control of f ne systems in simplices [9], we exhustively check for existence of controllers mking tringle n invrint, driving ll initil sttes in tringle to neighbor, two neighbors, nd three neighbors. While drmticlly 7 q q 6 q 7 q 7 67 q q 5 q
6 reducing the conservtiveness of the pproch, it induces non-deterministic trnsition system. The obtined trnsition system T is shown in Fig. 3, where the lbels (inputs) stnd for f ne feedbck controllers with n obvious mening. Note tht, motivted by the prticulr form of the formul (which is rechbility type formul) we do not consider the controllers corresponding to invrince, nd, therefore, T does not hve ny self trnsitions. By using the pproch from Section IV, we conclude tht the formul φ cn be stis ed by trjectories strting from ny tringle q i,i = 1,...,8. The LTL formul φ ccepts deterministic Büchi genertor B with 4 sttes, nd the constructed trnsition system T g hs 11 sttes. The winning strtegy W hs 4 sttes nd it is utomticlly generted, together with the set Q 0, fter providing the inputs T nd B. When implementing the (discrete) controller to the continuous system [19], discrete trnsition in T tkes plce when the current tringle is left, nd ech input to be pplied to T is mpped to n f ne feedbck controller s described in [9], pplied s long s the continuous trjectory evolves in the current tringle. The most interesting cse is tht of trjectories inititing in tringle q 3, becuse its corresponding discrete stte hs only non-deterministic outgoing trnsitions. We show two continuous trjectories strting in region q 3 nd corresponding to the winning strtegy we found s in Section IV. Even though the continuous trjectories rech different sequences of tringles, they both stisfy the formul. Only the rst prt of ech trjectory is shown (becuse the trjectories re in nite); fter region q 7 is hit, the trjectories will keep oscillting between tringles q 7 nd q 2. Solving the sme problem by ignoring the nondeterministic trnsitions of T nd using the pproch from Remrk 2, one would obtin tht the formul could be stis ed by strting from ny tringle except q 3 (note tht stte q 3 doesn t hve deterministic outgoing trnsitions in T ). By using the method from Remrk 3, the dversry would hve control on ll the outgoing trnsitions from sttes q 2,q 3,q 4,q 5, nd winning strtegies exist only when strting from sttes q 6,q 7,q 8. In the lter cse, q 7 will be rst visited, nd then q 4 (if q 4 ws visited rst, smrt dversry would keep the ply between sttes q 4 nd q 3, nd thus the gme would be lost). VI. CONCLUSIONS In this pper, we developed control strtegy for nondeterministic trnsition from speci ction given s n LTL formul over its set of sttes. The method is bsed on LTL gmes, nd consists of three steps: the construction of new trnsition system, the solution of n LTL gme, nd the genertion of the control strtegy for the initil system. The problem is motivted by nd pplied to plnning nd control of continuous systems from symbolic speci ctions given s temporl logic formuls. VII. ACKNOWLEDGMENTS This work ws prtilly supported by NSF CAREER nd NSF t Boston University. The uthors wish to thnk Pulo Tbud for useful discussions on this topic. REFERENCES [1] J. C. Ltombe, Robot Motion Plnning. Kluger Acdemic Pub., [2] H. Choset, K. M. Lynch, S. Hutchinson, G. Kntor, W. Burgrd, L. E. Kvrki, nd S. Thrun, Principles of Robot Motion: Theory, Algorithms, nd Implementtions. MIT Press, Boston, [3] S. Lindemnn nd S. LVlle, Computing smooth feedbck plns over cylindricl lgebric decompositions, in Proceedings of Robotics: Science nd Systems, Cmbridge, USA, June [4] D. Conner, H. Choset, nd A. Rizzi, Integrted plnning nd control for convex-bodied nonholonomic systems using locl feedbck control policies, in Proceedings of Robotics: Science nd Systems, Cmbridge, USA, June [5] C. Belt, V. Isler, nd G. J. Ppps, Discrete bstrctions for robot plnning nd control in polygonl environments, IEEE Trns. on Robotics, vol. 21, no. 5, pp , [6] M. Kloetzer nd C. Belt, A frmework for utomtic deployment of robots in 2D nd 3D environments, in IEEE/RSJ Interntionl Conference on Intelligent Robots nd Systems, Beijing, Chin, [7] G. E. Finekos, H. Kress-Gzit, nd G. J. Ppps, Hybrid controllers for pth plnning: temporl logic pproch, in Proceedings of the 2005 IEEE Conference on Decision nd Control, Seville, Spin, December [8] M. Kloetzer nd C. Belt, Hierrchicl bstrctions for robotic swrms, in IEEE Interntionl Conference on Robotics nd Automtion, Orlndo, FL, [9] L. Hbets, P. Collins, nd J. vn Schuppen, Rechbility nd control synthesis for piecewise-f ne hybrid systems on simplices, IEEE Trns. Aut. Control, vol. 51, pp , [10] L. Hbets, M. Kloetzer, nd C. Belt, Control of rectngulr multif ne hybrid systems, in 45th IEEE Conference on Decision nd Control, Sn Diego, CA, [11] W. Thoms, In nite gmes nd veri ction, in Proceedings of the Interntionl Conference on Computer Aided Veri ction, CAV 02, 2002, pp [12] R. Kumr nd V. K. Grg, Modeling nd Control of Logicl Discrete Event Systems. Boston, MA: Kluwer, [13] W. Thoms, Automt on in nite objects, in Hndbook of Theoreticl Computer Science, Volume B: Forml Models nd Semtics, J. vn Leeuwen, Ed. Amsterdm: Elsevier, 1990, pp [14] J. R. Büchi, On decision method in restricted second order rithmetic, in Proceedings of the Interntionl Congress on Logic, Methodology nd Philosophy of Science 1960, E. N. et l., Ed. Stnford, CA: Stnford University Press, 1962, p [15] S. L. Torre, Deciding gmes in LTL frgments, Seminr - Centre Fédéré en Véri ction, Belgium, Mrch [16] W. Thoms, Automt theoretic foundtions of in nite gmes, Spring School on In nite Gmes nd Their Applictions, Bonn, Mrch [17] S. Sfr, Complexity of utomt on in nite objects, Ph.D. disserttion, The Weizmn Institute of Science, Rehovot, Isrel, [18] G. Holzmnn, The Spin Model Checker, Primer nd Reference Mnul. Reding, Msschusetts: Addison-Wesley, [19] M. Kloetzer nd C. Belt, A fully utomted frmework for control of liner systems from LTL speci ctions, in The 9th Interntionl Workshop on Hybrid Systems: Computtion nd Control, 2006, pp [20], Mnging non-determinism in symbolic robot motion plnning nd control, Boston University, Tech. Rep., 2006, http//isi.bu.edu/ techrep/ltl-nondet.pdf. [21] A. Pnueli nd R. Rosner, On the synthesis of rective module, in Proceedings of the 16th ACM Symposium on Principles of Progrmming Lnguges, 1989, pp [22] R. Alur nd S. L. Torre, Deterministic genertors nd gmes for LTL frgments, in 16th IEEE Symposium on Logic in Computer Science, LICS 01, 2001, pp
Lecture 9: LTL and Büchi Automata
Lecture 9: LTL nd Büchi Automt 1 LTL Property Ptterns Quite often the requirements of system follow some simple ptterns. Sometimes we wnt to specify tht property should only hold in certin context, clled
More informationConvert the NFA into DFA
Convert the NF into F For ech NF we cn find F ccepting the sme lnguge. The numer of sttes of the F could e exponentil in the numer of sttes of the NF, ut in prctice this worst cse occurs rrely. lgorithm:
More informationFinite Automata. Informatics 2A: Lecture 3. John Longley. 22 September School of Informatics University of Edinburgh
Lnguges nd Automt Finite Automt Informtics 2A: Lecture 3 John Longley School of Informtics University of Edinburgh jrl@inf.ed.c.uk 22 September 2017 1 / 30 Lnguges nd Automt 1 Lnguges nd Automt Wht is
More informationBisimulation. R.J. van Glabbeek
Bisimultion R.J. vn Glbbeek NICTA, Sydney, Austrli. School of Computer Science nd Engineering, The University of New South Wles, Sydney, Austrli. Computer Science Deprtment, Stnford University, CA 94305-9045,
More information1 Nondeterministic Finite Automata
1 Nondeterministic Finite Automt Suppose in life, whenever you hd choice, you could try oth possiilities nd live your life. At the end, you would go ck nd choose the one tht worked out the est. Then you
More informationStrong Bisimulation. Overview. References. Actions Labeled transition system Transition semantics Simulation Bisimulation
Strong Bisimultion Overview Actions Lbeled trnsition system Trnsition semntics Simultion Bisimultion References Robin Milner, Communiction nd Concurrency Robin Milner, Communicting nd Mobil Systems 32
More informationCS:4330 Theory of Computation Spring Regular Languages. Equivalences between Finite automata and REs. Haniel Barbosa
CS:4330 Theory of Computtion Spring 208 Regulr Lnguges Equivlences between Finite utomt nd REs Hniel Brbos Redings for this lecture Chpter of [Sipser 996], 3rd edition. Section.3. Finite utomt nd regulr
More informationKNOWLEDGE-BASED AGENTS INFERENCE
AGENTS THAT REASON LOGICALLY KNOWLEDGE-BASED AGENTS Two components: knowledge bse, nd n inference engine. Declrtive pproch to building n gent. We tell it wht it needs to know, nd It cn sk itself wht to
More informationFinite Automata Theory and Formal Languages TMV027/DIT321 LP4 2018
Finite Automt Theory nd Forml Lnguges TMV027/DIT321 LP4 2018 Lecture 10 An Bove April 23rd 2018 Recp: Regulr Lnguges We cn convert between FA nd RE; Hence both FA nd RE ccept/generte regulr lnguges; More
More informationGood-for-Games Automata versus Deterministic Automata.
Good-for-Gmes Automt versus Deterministic Automt. Denis Kuperberg 1,2 Mich l Skrzypczk 1 1 University of Wrsw 2 IRIT/ONERA (Toulouse) Séminire MoVe 12/02/2015 LIF, Luminy Introduction Deterministic utomt
More informationHow to simulate Turing machines by invertible one-dimensional cellular automata
How to simulte Turing mchines by invertible one-dimensionl cellulr utomt Jen-Christophe Dubcq Déprtement de Mthémtiques et d Informtique, École Normle Supérieure de Lyon, 46, llée d Itlie, 69364 Lyon Cedex
More informationNondeterminism and Nodeterministic Automata
Nondeterminism nd Nodeterministic Automt 61 Nondeterminism nd Nondeterministic Automt The computtionl mchine models tht we lerned in the clss re deterministic in the sense tht the next move is uniquely
More informationReinforcement Learning
Reinforcement Lerning Tom Mitchell, Mchine Lerning, chpter 13 Outline Introduction Comprison with inductive lerning Mrkov Decision Processes: the model Optiml policy: The tsk Q Lerning: Q function Algorithm
More informationCSCI 340: Computational Models. Kleene s Theorem. Department of Computer Science
CSCI 340: Computtionl Models Kleene s Theorem Chpter 7 Deprtment of Computer Science Unifiction In 1954, Kleene presented (nd proved) theorem which (in our version) sttes tht if lnguge cn e defined y ny
More informationTheory of Computation Regular Languages. (NTU EE) Regular Languages Fall / 38
Theory of Computtion Regulr Lnguges (NTU EE) Regulr Lnguges Fll 2017 1 / 38 Schemtic of Finite Automt control 0 0 1 0 1 1 1 0 Figure: Schemtic of Finite Automt A finite utomton hs finite set of control
More informationThe Regulated and Riemann Integrals
Chpter 1 The Regulted nd Riemnn Integrls 1.1 Introduction We will consider severl different pproches to defining the definite integrl f(x) dx of function f(x). These definitions will ll ssign the sme vlue
More informationTheory of Computation Regular Languages
Theory of Computtion Regulr Lnguges Bow-Yw Wng Acdemi Sinic Spring 2012 Bow-Yw Wng (Acdemi Sinic) Regulr Lnguges Spring 2012 1 / 38 Schemtic of Finite Automt control 0 0 1 0 1 1 1 0 Figure: Schemtic of
More information1.4 Nonregular Languages
74 1.4 Nonregulr Lnguges The number of forml lnguges over ny lphbet (= decision/recognition problems) is uncountble On the other hnd, the number of regulr expressions (= strings) is countble Hence, ll
More informationNFAs and Regular Expressions. NFA-ε, continued. Recall. Last class: Today: Fun:
CMPU 240 Lnguge Theory nd Computtion Spring 2019 NFAs nd Regulr Expressions Lst clss: Introduced nondeterministic finite utomt with -trnsitions Tody: Prove n NFA- is no more powerful thn n NFA Introduce
More informationCMSC 330: Organization of Programming Languages. DFAs, and NFAs, and Regexps (Oh my!)
CMSC 330: Orgniztion of Progrmming Lnguges DFAs, nd NFAs, nd Regexps (Oh my!) CMSC330 Spring 2018 Types of Finite Automt Deterministic Finite Automt (DFA) Exctly one sequence of steps for ech string All
More informationCS 275 Automata and Formal Language Theory
CS 275 Automt nd Forml Lnguge Theory Course Notes Prt II: The Recognition Problem (II) Chpter II.6.: Push Down Automt Remrk: This mteril is no longer tught nd not directly exm relevnt Anton Setzer (Bsed
More informationReinforcement learning II
CS 1675 Introduction to Mchine Lerning Lecture 26 Reinforcement lerning II Milos Huskrecht milos@cs.pitt.edu 5329 Sennott Squre Reinforcement lerning Bsics: Input x Lerner Output Reinforcement r Critic
More informationCS103B Handout 18 Winter 2007 February 28, 2007 Finite Automata
CS103B ndout 18 Winter 2007 Ferury 28, 2007 Finite Automt Initil text y Mggie Johnson. Introduction Severl childrens gmes fit the following description: Pieces re set up on plying ord; dice re thrown or
More informationCoalgebra, Lecture 15: Equations for Deterministic Automata
Colger, Lecture 15: Equtions for Deterministic Automt Julin Slmnc (nd Jurrin Rot) Decemer 19, 2016 In this lecture, we will study the concept of equtions for deterministic utomt. The notes re self contined
More informationFinite Automata. Informatics 2A: Lecture 3. Mary Cryan. 21 September School of Informatics University of Edinburgh
Finite Automt Informtics 2A: Lecture 3 Mry Cryn School of Informtics University of Edinburgh mcryn@inf.ed.c.uk 21 September 2018 1 / 30 Lnguges nd Automt Wht is lnguge? Finite utomt: recp Some forml definitions
More information12.1 Nondeterminism Nondeterministic Finite Automata. a a b ε. CS125 Lecture 12 Fall 2016
CS125 Lecture 12 Fll 2016 12.1 Nondeterminism The ide of nondeterministic computtions is to llow our lgorithms to mke guesses, nd only require tht they ccept when the guesses re correct. For exmple, simple
More informationNon-Deterministic Finite Automata. Fall 2018 Costas Busch - RPI 1
Non-Deterministic Finite Automt Fll 2018 Costs Busch - RPI 1 Nondeterministic Finite Automton (NFA) Alphbet ={} q q2 1 q 0 q 3 Fll 2018 Costs Busch - RPI 2 Nondeterministic Finite Automton (NFA) Alphbet
More informationThis lecture covers Chapter 8 of HMU: Properties of CFLs
This lecture covers Chpter 8 of HMU: Properties of CFLs Turing Mchine Extensions of Turing Mchines Restrictions of Turing Mchines Additionl Reding: Chpter 8 of HMU. Turing Mchine: Informl Definition B
More information19 Optimal behavior: Game theory
Intro. to Artificil Intelligence: Dle Schuurmns, Relu Ptrscu 1 19 Optiml behvior: Gme theory Adversril stte dynmics hve to ccount for worst cse Compute policy π : S A tht mximizes minimum rewrd Let S (,
More informationCMSC 330: Organization of Programming Languages
CMSC 330: Orgniztion of Progrmming Lnguges Finite Automt 2 CMSC 330 1 Types of Finite Automt Deterministic Finite Automt (DFA) Exctly one sequence of steps for ech string All exmples so fr Nondeterministic
More informationCS 275 Automata and Formal Language Theory
CS 275 Automt nd Forml Lnguge Theory Course Notes Prt II: The Recognition Problem (II) Chpter II.5.: Properties of Context Free Grmmrs (14) Anton Setzer (Bsed on book drft by J. V. Tucker nd K. Stephenson)
More informationHandout: Natural deduction for first order logic
MATH 457 Introduction to Mthemticl Logic Spring 2016 Dr Json Rute Hndout: Nturl deduction for first order logic We will extend our nturl deduction rules for sententil logic to first order logic These notes
More informationChapter Five: Nondeterministic Finite Automata. Formal Language, chapter 5, slide 1
Chpter Five: Nondeterministic Finite Automt Forml Lnguge, chpter 5, slide 1 1 A DFA hs exctly one trnsition from every stte on every symol in the lphet. By relxing this requirement we get relted ut more
More informationModel Reduction of Finite State Machines by Contraction
Model Reduction of Finite Stte Mchines y Contrction Alessndro Giu Dip. di Ingegneri Elettric ed Elettronic, Università di Cgliri, Pizz d Armi, 09123 Cgliri, Itly Phone: +39-070-675-5892 Fx: +39-070-675-5900
More informationTypes of Finite Automata. CMSC 330: Organization of Programming Languages. Comparing DFAs and NFAs. Comparing DFAs and NFAs (cont.) Finite Automata 2
CMSC 330: Orgniztion of Progrmming Lnguges Finite Automt 2 Types of Finite Automt Deterministic Finite Automt () Exctly one sequence of steps for ech string All exmples so fr Nondeterministic Finite Automt
More informationTypes of Finite Automata. CMSC 330: Organization of Programming Languages. Comparing DFAs and NFAs. NFA for (a b)*abb.
CMSC 330: Orgniztion of Progrmming Lnguges Finite Automt 2 Types of Finite Automt Deterministic Finite Automt () Exctly one sequence of steps for ech string All exmples so fr Nondeterministic Finite Automt
More informationMore on automata. Michael George. March 24 April 7, 2014
More on utomt Michel George Mrch 24 April 7, 2014 1 Automt constructions Now tht we hve forml model of mchine, it is useful to mke some generl constructions. 1.1 DFA Union / Product construction Suppose
More informationNon Deterministic Automata. Linz: Nondeterministic Finite Accepters, page 51
Non Deterministic Automt Linz: Nondeterministic Finite Accepters, pge 51 1 Nondeterministic Finite Accepter (NFA) Alphbet ={} q 1 q2 q 0 q 3 2 Nondeterministic Finite Accepter (NFA) Alphbet ={} Two choices
More informationOn Determinisation of History-Deterministic Automata.
On Deterministion of History-Deterministic Automt. Denis Kupererg Mich l Skrzypczk University of Wrsw YR-ICALP 2014 Copenhgen Introduction Deterministic utomt re centrl tool in utomt theory: Polynomil
More information12.1 Nondeterminism Nondeterministic Finite Automata. a a b ε. CS125 Lecture 12 Fall 2014
CS125 Lecture 12 Fll 2014 12.1 Nondeterminism The ide of nondeterministic computtions is to llow our lgorithms to mke guesses, nd only require tht they ccept when the guesses re correct. For exmple, simple
More informationFormal Methods in Software Engineering
Forml Methods in Softwre Engineering Lecture 09 orgniztionl issues Prof. Dr. Joel Greenyer Decemer 9, 2014 Written Exm The written exm will tke plce on Mrch 4 th, 2015 The exm will tke 60 minutes nd strt
More informationMinimal DFA. minimal DFA for L starting from any other
Miniml DFA Among the mny DFAs ccepting the sme regulr lnguge L, there is exctly one (up to renming of sttes) which hs the smllest possile numer of sttes. Moreover, it is possile to otin tht miniml DFA
More informationAnatomy of a Deterministic Finite Automaton. Deterministic Finite Automata. A machine so simple that you can understand it in less than one minute
Victor Admchik Dnny Sletor Gret Theoreticl Ides In Computer Science CS 5-25 Spring 2 Lecture 2 Mr 3, 2 Crnegie Mellon University Deterministic Finite Automt Finite Automt A mchine so simple tht you cn
More informationNew Expansion and Infinite Series
Interntionl Mthemticl Forum, Vol. 9, 204, no. 22, 06-073 HIKARI Ltd, www.m-hikri.com http://dx.doi.org/0.2988/imf.204.4502 New Expnsion nd Infinite Series Diyun Zhng College of Computer Nnjing University
More informationCS415 Compilers. Lexical Analysis and. These slides are based on slides copyrighted by Keith Cooper, Ken Kennedy & Linda Torczon at Rice University
CS415 Compilers Lexicl Anlysis nd These slides re sed on slides copyrighted y Keith Cooper, Ken Kennedy & Lind Torczon t Rice University First Progrmming Project Instruction Scheduling Project hs een posted
More informationLecture 09: Myhill-Nerode Theorem
CS 373: Theory of Computtion Mdhusudn Prthsrthy Lecture 09: Myhill-Nerode Theorem 16 Ferury 2010 In this lecture, we will see tht every lnguge hs unique miniml DFA We will see this fct from two perspectives
More informationDuality # Second iteration for HW problem. Recall our LP example problem we have been working on, in equality form, is given below.
Dulity #. Second itertion for HW problem Recll our LP emple problem we hve been working on, in equlity form, is given below.,,,, 8 m F which, when written in slightly different form, is 8 F Recll tht we
More informationCS5371 Theory of Computation. Lecture 20: Complexity V (Polynomial-Time Reducibility)
CS5371 Theory of Computtion Lecture 20: Complexity V (Polynomil-Time Reducibility) Objectives Polynomil Time Reducibility Prove Cook-Levin Theorem Polynomil Time Reducibility Previously, we lernt tht if
More informationChapter 4 Contravariance, Covariance, and Spacetime Diagrams
Chpter 4 Contrvrince, Covrince, nd Spcetime Digrms 4. The Components of Vector in Skewed Coordintes We hve seen in Chpter 3; figure 3.9, tht in order to show inertil motion tht is consistent with the Lorentz
More informationC. C^mpenu, K. Slom, S. Yu upper boun of mn. So our result is tight only for incomplete DF's. For restricte vlues of m n n we present exmples of DF's
Journl of utomt, Lnguges n Combintorics u (v) w, x{y c OttovonGuerickeUniversitt Mgeburg Tight lower boun for the stte complexity of shue of regulr lnguges Cezr C^mpenu, Ki Slom Computing n Informtion
More informationRegular expressions, Finite Automata, transition graphs are all the same!!
CSI 3104 /Winter 2011: Introduction to Forml Lnguges Chpter 7: Kleene s Theorem Chpter 7: Kleene s Theorem Regulr expressions, Finite Automt, trnsition grphs re ll the sme!! Dr. Neji Zgui CSI3104-W11 1
More informationImproper Integrals, and Differential Equations
Improper Integrls, nd Differentil Equtions October 22, 204 5.3 Improper Integrls Previously, we discussed how integrls correspond to res. More specificlly, we sid tht for function f(x), the region creted
More informationCS 373, Spring Solutions to Mock midterm 1 (Based on first midterm in CS 273, Fall 2008.)
CS 373, Spring 29. Solutions to Mock midterm (sed on first midterm in CS 273, Fll 28.) Prolem : Short nswer (8 points) The nswers to these prolems should e short nd not complicted. () If n NF M ccepts
More informationFormal languages, automata, and theory of computation
Mälrdlen University TEN1 DVA337 2015 School of Innovtion, Design nd Engineering Forml lnguges, utomt, nd theory of computtion Thursdy, Novemer 5, 14:10-18:30 Techer: Dniel Hedin, phone 021-107052 The exm
More informationAutomata and Languages
Automt nd Lnguges Prof. Mohmed Hmd Softwre Engineering Lb. The University of Aizu Jpn Grmmr Regulr Grmmr Context-free Grmmr Context-sensitive Grmmr Regulr Lnguges Context Free Lnguges Context Sensitive
More informationRiemann Sums and Riemann Integrals
Riemnn Sums nd Riemnn Integrls Jmes K. Peterson Deprtment of Biologicl Sciences nd Deprtment of Mthemticl Sciences Clemson University August 26, 203 Outline Riemnn Sums Riemnn Integrls Properties Abstrct
More informationAn Automata Toolbox. Mikołaj Bojańczyk and Wojciech Czerwiński
An Automt Toolbox Mikołj Bojńczyk nd Wojciech Czerwiński Prefce T hese re lecture notes for course on dvnced utomt theory, tht we gve t the University of Wrsw in the yers 2015-2018. The mteril ws chosen
More informationTHE EXISTENCE-UNIQUENESS THEOREM FOR FIRST-ORDER DIFFERENTIAL EQUATIONS.
THE EXISTENCE-UNIQUENESS THEOREM FOR FIRST-ORDER DIFFERENTIAL EQUATIONS RADON ROSBOROUGH https://intuitiveexplntionscom/picrd-lindelof-theorem/ This document is proof of the existence-uniqueness theorem
More informationRiemann Sums and Riemann Integrals
Riemnn Sums nd Riemnn Integrls Jmes K. Peterson Deprtment of Biologicl Sciences nd Deprtment of Mthemticl Sciences Clemson University August 26, 2013 Outline 1 Riemnn Sums 2 Riemnn Integrls 3 Properties
More informationTheoretical foundations of Gaussian quadrature
Theoreticl foundtions of Gussin qudrture 1 Inner product vector spce Definition 1. A vector spce (or liner spce) is set V = {u, v, w,...} in which the following two opertions re defined: (A) Addition of
More informationFinite Automata-cont d
Automt Theory nd Forml Lnguges Professor Leslie Lnder Lecture # 6 Finite Automt-cont d The Pumping Lemm WEB SITE: http://ingwe.inghmton.edu/ ~lnder/cs573.html Septemer 18, 2000 Exmple 1 Consider L = {ww
More informationCHAPTER 1 Regular Languages. Contents
Finite Automt (FA or DFA) CHAPTE 1 egulr Lnguges Contents definitions, exmples, designing, regulr opertions Non-deterministic Finite Automt (NFA) definitions, euivlence of NFAs nd DFAs, closure under regulr
More informationAutomata, Games, and Verification
Automt, Gmes, nd Verifiction Prof. Bernd Finkbeiner, Ph.D. Srlnd University Summer Term 2015 Lecture Notes by Bernd Finkbeiner, Felix Klein, Tobis Slzmnn These lecture notes re working document nd my contin
More information20 MATHEMATICS POLYNOMIALS
0 MATHEMATICS POLYNOMIALS.1 Introduction In Clss IX, you hve studied polynomils in one vrible nd their degrees. Recll tht if p(x) is polynomil in x, the highest power of x in p(x) is clled the degree of
More informationSection 6.1 INTRO to LAPLACE TRANSFORMS
Section 6. INTRO to LAPLACE TRANSFORMS Key terms: Improper Integrl; diverge, converge A A f(t)dt lim f(t)dt Piecewise Continuous Function; jump discontinuity Function of Exponentil Order Lplce Trnsform
More informationChapter 14. Matrix Representations of Linear Transformations
Chpter 4 Mtrix Representtions of Liner Trnsformtions When considering the Het Stte Evolution, we found tht we could describe this process using multipliction by mtrix. This ws nice becuse computers cn
More informationMAA 4212 Improper Integrals
Notes by Dvid Groisser, Copyright c 1995; revised 2002, 2009, 2014 MAA 4212 Improper Integrls The Riemnn integrl, while perfectly well-defined, is too restrictive for mny purposes; there re functions which
More informationAUTOMATA AND LANGUAGES. Definition 1.5: Finite Automaton
25. Finite Automt AUTOMATA AND LANGUAGES A system of computtion tht only hs finite numer of possile sttes cn e modeled using finite utomton A finite utomton is often illustrted s stte digrm d d d. d q
More informationIntermediate Math Circles Wednesday, November 14, 2018 Finite Automata II. Nickolas Rollick a b b. a b 4
Intermedite Mth Circles Wednesdy, Novemer 14, 2018 Finite Automt II Nickols Rollick nrollick@uwterloo.c Regulr Lnguges Lst time, we were introduced to the ide of DFA (deterministic finite utomton), one
More informationDesigning finite automata II
Designing finite utomt II Prolem: Design DFA A such tht L(A) consists of ll strings of nd which re of length 3n, for n = 0, 1, 2, (1) Determine wht to rememer out the input string Assign stte to ech of
More informationChapter 2 Finite Automata
Chpter 2 Finite Automt 28 2.1 Introduction Finite utomt: first model of the notion of effective procedure. (They lso hve mny other pplictions). The concept of finite utomton cn e derived y exmining wht
More information1. For each of the following theorems, give a two or three sentence sketch of how the proof goes or why it is not true.
York University CSE 2 Unit 3. DFA Clsses Converting etween DFA, NFA, Regulr Expressions, nd Extended Regulr Expressions Instructor: Jeff Edmonds Don t chet y looking t these nswers premturely.. For ech
More informationCS 267: Automated Verification. Lecture 8: Automata Theoretic Model Checking. Instructor: Tevfik Bultan
CS 267: Automted Verifiction Lecture 8: Automt Theoretic Model Checking Instructor: Tevfik Bultn LTL Properties Büchi utomt [Vrdi nd Wolper LICS 86] Büchi utomt: Finite stte utomt tht ccept infinite strings
More informationLearning Moore Machines from Input-Output Traces
Lerning Moore Mchines from Input-Output Trces Georgios Gintmidis 1 nd Stvros Tripkis 1,2 1 Alto University, Finlnd 2 UC Berkeley, USA Motivtion: lerning models from blck boxes Inputs? Lerner Forml Model
More informationp-adic Egyptian Fractions
p-adic Egyptin Frctions Contents 1 Introduction 1 2 Trditionl Egyptin Frctions nd Greedy Algorithm 2 3 Set-up 3 4 p-greedy Algorithm 5 5 p-egyptin Trditionl 10 6 Conclusion 1 Introduction An Egyptin frction
More information5.1 Definitions and Examples 5.2 Deterministic Pushdown Automata
CSC4510 AUTOMATA 5.1 Definitions nd Exmples 5.2 Deterministic Pushdown Automt Definitions nd Exmples A lnguge cn be generted by CFG if nd only if it cn be ccepted by pushdown utomton. A pushdown utomton
More informationRecitation 3: More Applications of the Derivative
Mth 1c TA: Pdric Brtlett Recittion 3: More Applictions of the Derivtive Week 3 Cltech 2012 1 Rndom Question Question 1 A grph consists of the following: A set V of vertices. A set E of edges where ech
More informationLTL Translation Improvements in Spot
LTL Trnsltion Improvements in Spot Alexndre Duret-Lutz http://www.lrde.epit.fr/~dl/ VECoS'11 16 September 2011 Alexndre Duret-Lutz LTL Trnsltion Improvements 1 / 19 Context High-level
More informationSoftware Engineering using Formal Methods
Softwre Engineering using Forml Methods Propositionl nd (Liner) Temporl Logic Wolfgng Ahrendt 13th Septemer 2016 SEFM: Liner Temporl Logic /GU 160913 1 / 60 Recpitultion: FormlistionFormlistion: Syntx,
More informationAdministrivia CSE 190: Reinforcement Learning: An Introduction
Administrivi CSE 190: Reinforcement Lerning: An Introduction Any emil sent to me bout the course should hve CSE 190 in the subject line! Chpter 4: Dynmic Progrmming Acknowledgment: A good number of these
More information1.3 Regular Expressions
56 1.3 Regulr xpressions These hve n importnt role in describing ptterns in serching for strings in mny pplictions (e.g. wk, grep, Perl,...) All regulr expressions of lphbet re 1.Ønd re regulr expressions,
More informationRefined interfaces for compositional verification
Refined interfces for compositionl verifiction Frédéric Lng INRI Rhône-lpes http://www.inrilpes.fr/vsy Motivtion Enumertive verifiction of concurrent systems Prllel composition of synchronous processes
More informationState space systems analysis (continued) Stability. A. Definitions A system is said to be Asymptotically Stable (AS) when it satisfies
Stte spce systems nlysis (continued) Stbility A. Definitions A system is sid to be Asymptoticlly Stble (AS) when it stisfies ut () = 0, t > 0 lim xt () 0. t A system is AS if nd only if the impulse response
More informationKleene Theorems for Free Choice Nets Labelled with Distributed Alphabets
Kleene Theorems for Free Choice Nets Lbelled with Distributed Alphbets Rmchndr Phwde Indin Institute of Technology Dhrwd, Dhrwd 580011, Indi Emil: prb@iitdh.c.in Abstrct. We provided [15] expressions for
More informationParse trees, ambiguity, and Chomsky normal form
Prse trees, miguity, nd Chomsky norml form In this lecture we will discuss few importnt notions connected with contextfree grmmrs, including prse trees, miguity, nd specil form for context-free grmmrs
More informationReview of basic calculus
Review of bsic clculus This brief review reclls some of the most importnt concepts, definitions, nd theorems from bsic clculus. It is not intended to tech bsic clculus from scrtch. If ny of the items below
More informationLecture 1. Functional series. Pointwise and uniform convergence.
1 Introduction. Lecture 1. Functionl series. Pointwise nd uniform convergence. In this course we study mongst other things Fourier series. The Fourier series for periodic function f(x) with period 2π is
More informationMAC-solutions of the nonexistent solutions of mathematical physics
Proceedings of the 4th WSEAS Interntionl Conference on Finite Differences - Finite Elements - Finite Volumes - Boundry Elements MAC-solutions of the nonexistent solutions of mthemticl physics IGO NEYGEBAUE
More informationAutomata Theory 101. Introduction. Outline. Introduction Finite Automata Regular Expressions ω-automata. Ralf Huuck.
Outline Automt Theory 101 Rlf Huuck Introduction Finite Automt Regulr Expressions ω-automt Session 1 2006 Rlf Huuck 1 Session 1 2006 Rlf Huuck 2 Acknowledgement Some slides re sed on Wolfgng Thoms excellent
More informationFormal Languages and Automata
Moile Computing nd Softwre Engineering p. 1/5 Forml Lnguges nd Automt Chpter 2 Finite Automt Chun-Ming Liu cmliu@csie.ntut.edu.tw Deprtment of Computer Science nd Informtion Engineering Ntionl Tipei University
More informationUnit #9 : Definite Integral Properties; Fundamental Theorem of Calculus
Unit #9 : Definite Integrl Properties; Fundmentl Theorem of Clculus Gols: Identify properties of definite integrls Define odd nd even functions, nd reltionship to integrl vlues Introduce the Fundmentl
More informationCS 330 Formal Methods and Models Dana Richards, George Mason University, Spring 2016 Quiz Solutions
CS 330 Forml Methods nd Models Dn Richrds, George Mson University, Spring 2016 Quiz Solutions Quiz 1, Propositionl Logic Dte: Ferury 9 1. (4pts) ((p q) (q r)) (p r), prove tutology using truth tles. p
More information7.2 The Definite Integral
7.2 The Definite Integrl the definite integrl In the previous section, it ws found tht if function f is continuous nd nonnegtive, then the re under the grph of f on [, b] is given by F (b) F (), where
More informationNFA DFA Example 3 CMSC 330: Organization of Programming Languages. Equivalence of DFAs and NFAs. Equivalence of DFAs and NFAs (cont.
NFA DFA Exmple 3 CMSC 330: Orgniztion of Progrmming Lnguges NFA {B,D,E {A,E {C,D {E Finite Automt, con't. R = { {A,E, {B,D,E, {C,D, {E 2 Equivlence of DFAs nd NFAs Any string from {A to either {D or {CD
More informationDIRECT CURRENT CIRCUITS
DRECT CURRENT CUTS ELECTRC POWER Consider the circuit shown in the Figure where bttery is connected to resistor R. A positive chrge dq will gin potentil energy s it moves from point to point b through
More informationReview of Gaussian Quadrature method
Review of Gussin Qudrture method Nsser M. Asi Spring 006 compiled on Sundy Decemer 1, 017 t 09:1 PM 1 The prolem To find numericl vlue for the integrl of rel vlued function of rel vrile over specific rnge
More informationCMPSCI 250: Introduction to Computation. Lecture #31: What DFA s Can and Can t Do David Mix Barrington 9 April 2014
CMPSCI 250: Introduction to Computtion Lecture #31: Wht DFA s Cn nd Cn t Do Dvid Mix Brrington 9 April 2014 Wht DFA s Cn nd Cn t Do Deterministic Finite Automt Forml Definition of DFA s Exmples of DFA
More informationBernoulli Numbers Jeff Morton
Bernoulli Numbers Jeff Morton. We re interested in the opertor e t k d k t k, which is to sy k tk. Applying this to some function f E to get e t f d k k tk d k f f + d k k tk dk f, we note tht since f
More informationPavel Rytí. November 22, 2011 Discrete Math Seminar - Simon Fraser University
Geometric representtions of liner codes Pvel Rytí Deprtment of Applied Mthemtics Chrles University in Prgue Advisor: Mrtin Loebl November, 011 Discrete Mth Seminr - Simon Frser University Bckground Liner
More informationTime Optimal Control of the Brockett Integrator
Milno (Itly) August 8 - September, 011 Time Optiml Control of the Brockett Integrtor S. Sinh Deprtment of Mthemtics, IIT Bomby, Mumbi, Indi (emil : sunnysphs4891@gmil.com) Abstrct: The Brockett integrtor
More information