Cooperative Collision Avoidance between Multiple Mobile Robots
|
|
- Rosalyn Lynette Harper
- 5 years ago
- Views:
Transcription
1 Cooeratve Collson Avodance between Multle Moble Robots Atsush Fujmor,* Masato Teramoto Deartment of Mechancal Engneerng Shzuoka Unversty 3-5-1, Johoku, Hamamatsu , Jaan e-mal: Peter N. Nkforuk, Madan M. Guta Deartment of Mechancal Engneerng Unversty of Saskatchewan Saskatoon, Saskatchewan S7N 5A9, Canada Receved March 27, 1999; acceted February 10, 2000 Ths aer resents a new collson avodance technque, called cooeratve collson avodance, for multle moble robots. The detecton of the danger of collson between two moble robots s dscussed wth resect to the geometrc asects of ther aths as are cooeratve collson avodance behavors. The drecton control command and the velocty control command for the cooeratve collson avodance are then roosed. The avodance technque s extended to cases n whch the number of moble robots s more than two. Furthermore, the condtons for collson avodance are consdered wth resect to the navgaton arameters and gudelnes of desgnng the navgaton arameters are obtaned. The effectveness of the roosed technque s demonstrated by means of numercal smulaton and navgaton exerments usng two real moble robots named Poneer John Wley & Sons, Inc. I. INTRODUCTION Moble robots have been the subject of much research and develoment due to ther alcablty n a wde varety of tasks n ndustry, human su- * To whom all corresondence should be addressed. orted works, and elsewhere. If the task requested cannot be easly carred out by a sngle robot, multle robots should be cooeratvely used. 1 4 Snce the use of multle robots may lead to collson, ther navgaton and collson avodance have been dscussed from varous onts of vew. In Ref. 1, for examle, a arallel rocessng system was con- () ( ) Journal of Robotc Systems 17 7, by John Wley & Sons, Inc.
2 348 Journal of Robotc Systems 2000 structed for the cellular robotc system and was aled to the behavor decson of multle moble robots. Matarc 5,6 roosed avodance behavor rules when a collson between two moble robots was redcted: stong the robots for a fxed erod or changng ther drectons. Arkn 7 and Sughara et al. 8 roosed algorthms combnng these rules. Shan and Hasegawa 9 roosed behavorbased moton lannng of multle moble robots n a narrow assage. On the other hand, learnng technques were ncororated nto neural networks andor fuzzy logc controllers for moble robot navgaton. 10,11 Iteratng navgaton of moble robots n a gven envronment, navgaton arameters assocated wth robot s behavors were adjusted by the learnng technques. These aroaches dd not gve analytcal dscussons for selectng the navgaton arameters although the effectveness was shown n smulaton and navgaton exerments. Therefore, the ntal selecton of the navgaton arameters remans to be solved. Ths aer resents the navgaton of multle moble robots wth collson avodance n a two-dmensonal free-sace envronment and dscusses collson avodance condtons from the theoretcal onts of vew. Each robot has ts own goal or oston to be navgated. The objectve s to navgate the multle moble robots to ther goals wthout collsons between them. To do ths, ths aer rooses a new collson avodance technque called cooeratve collson avodance and gves the avodance condtons wth resect to the navgaton arameters. The navgaton scheme used n ths aer s based on the adatve navgaton technque 12 n whch the navgaton law s gven by a frst-order dfferental equaton, and the navgaton of the robot to the goal and obstacle avodance are done by swtchng the drecton angle command adatvely. In ths aer, the cooeratve collson avodance behavors of the multle moble robots are then ncororated nto the adatve navgaton technque. The remander of ths aer s organzed as follows. The statements of roblem are resented n Secton II. The cooeratve collson avodance s dscussed n Secton III. The detecton of collson between two moble robots s frst dscussed wth resect to asects under whch the two robots encounter each other. The drecton control command and the velocty control command for the cooeratve collson avodance are then roosed. The avodance technque s extended to cases n whch the number of moble robots s more than two. The condtons for collson avodance are consdered wth resect to the navgaton arameters n Secton IV. Accordng to the condtons, a desgn rocedure of the navgaton arameters s gven. The effectveness of the roosed technque s demonstrated by means of numercal smulaton examles n Secton V. In Secton VI, the roosed technque s aled to an exermental robot named Poneer-1. The concluson s resented n Secton VII. II. STATEMENT OF PROBLEM Ths secton descrbes a navgaton roblem of multle moble robots n a two-dmensonal free-sace envronment. Let n be the number of moble robots. r The th moble robot s denoted as R and ts oston s reresented by the cartesan coordnates Žx Ž t., y Ž t.. where t s tme. The drecton angle of the th robot s Ž t., Ž Ž t. rad., whch s measured from the x-axs. The velocty of the th robot s v Ž t.. The equatons of moton of the nr moble robots are then gven by x Ž t. v Ž t. cos Ž t. Ž 1,..., n.. Ž 1. r y Ž t. v Ž t. sn Ž t. The objectve s to navgate the nr moble robots from ther startng onts S Ž 1,..., n. r to ther own goals G Ž 1,..., n. r wthout any collsons between them. The followng assumton s gven on the robots. ( ) A1 All of moble robots have the same sze and the same erformance. The radus of the robots s r. R ( A2) Žx Ž t., y Ž t., Ž t., v Ž t.. Ž 1,..., n. r are known to each other by some means of communcaton. The drecton angle of the moble robots s gven by Ž t. Ž t. Ž t. Ž 1,..., n., Ž 2. r where s a ostve constant. Ths rocess can be used not only to navgate moble robots to ther goal, but also to avod collsons by swtchng the drecton angle command Ž t.. 12 Equaton Ž. 2 s called the drecton control.
3 Fujmor et al.: Collson Avodance between Robots 349 Smlarly, the velocty of the moble robots s gven by v Ž t. v Ž t. v Ž t. Ž 1,..., n., Ž 3. v r where 0v v t v. 4 mn max s a ostve constant, v Ž. v t s the velocty com- mand, vmn s the mnmum velocty, and vmax s the maxmum velocty. Equaton Ž. 3 s called the velocty control. Based on the adatve navgaton technque, 12 the drecton angle command Ž. t and the velocty command v Ž. t are gven accordng to the follow- ng three modes: Fgure 1. Two moble robots. Navgaton mode: Navgate a robot to ts goal f no collson s detected and the robot s not at goal. Lettng Ž. t be the angle from the oston of the th robot R to ts goal G, Ž t. s gven by Ž t. Ž t., whle v Ž t. s gven by v Ž. t v 0, where v0 s the navga- ton velocty. Cooeratve avodance mode: Make a robot avod collsons wth other robots f collsons are redcted. Ž. t and v Ž. t wll be gven n the later sectons. Fnal mode: Aroach a robot to ts goal when the robot s near the goal. Let lg be the dstance between the th robot R and ts goal G. Near means that l d, where d s a G f f range to the fnal mode. When navgatng robot R to goal G, let t f be the tme n whch l becomes l d frst. Then, Ž t. s fxed G G f as Ž t. Ž t. for tt. v Ž. f f t s decreased to zero whle aroachng to the goal. control commands for the avodance. The cooeratve collson avodance between two moble robots s then extended to cases n whch the number of moble robots s more than two. A. Collson Detecton and Mode Selecton Fgure 1 shows a stuaton where the oston, drecton angle, and velocty of two moble robots R 1 and R are Žx Ž. t, y Ž.. t, Ž. t, and v Ž.Ž t 1, 2. 2, resectvely. If the robots face n the drecton of the dashed arrows, a collson wll not haen. However, f the robots face n the drecton of the sold arrows, the two moble robots may collde wth each other at a ont P 12, whch s called the redcted crossng ont. Let lr j be the dstance between robots R and R j. In Fg. 1, lr and dlr dt are gven by The detals of the navgaton mode and the fnal mode are descrbed n Ref. 12. The rest of ths aer therefore deals wth the cooeratve avodance mode. III. COOPERATIVE COLLISION AVOIDANCE Ths secton resents the collson avodance between multle moble robots, that s, the cooeratve avodance mode. Frst, a concet of the cooeratve collson avodance s descrbed wth two moble robots, that s, the collson detecton, the cooeratve collson avodance behavors, and the 2 2 R l 'Ž x x. Ž y y. Ž 5. dlr 12 1 Ž v cos v cos.ž x x dt l R12 v sn v sn y y Usng them, the danger of collson s classfed as Table 1. Evaluatng the danger of collson, t s seen that an avodance behavor s requred for Case 4. Thus, lr and dlr dt can be used as a swtch to the cooeratve avodance mode. Usng l G, l R, and j dl dt, the mode selecton of robot R s gven as Rj
4 350 Journal of Robotc Systems 2000 Table 1. Danger of collson wth resect to l and R dl dt. R Case lr dlrdt Danger of collson 1 long 0 Danger s small and s gong to be decreased. 2 long 0 Danger s small but s gong to be ncreased. 3 short 0 Danger s great but s gong to be decreased. 4 short 0 Danger s great and s gong to be ncreased. follows: f lg d f then fnal mode s selected. else f lr d and dlr dtd j j v then cooeratve avodance mode s selected. else navgaton mode s selected. end f d Ž 0. and d Ž 0. v are, resectvely, the dstance- and the dervatve-swtch to the cooeratve avodance mode to be desgned. B. Cooeratve Collson Avodance Behavors As descrbed n Secton II, moble robots have two control strateges, the drecton angle control and the velocty control for navgaton and collson avodance. The urose of ths secton s to resent an avodance technque usng these strateges for the cooeratve collson avodance. Let and Ž,. be defned as the crossng angle and the relatve angle resectvely as shown n Fg. 1. Usng the crossng angle, asects under whch two moble robots may encounter each other n a two-dmensonal free-sace envronment are classfed as gven n Fg. 2. In Fg. 2Ž. a, a collson wll occur unless the robots change ther drectons. In ths case, the drecton control s requred to avod collson. The velocty control, however, s of no use. In Fg. 2Ž. c, on the other hand, a collson may be avoded by changng the drectons of the two robots, or changng ther veloctes so that they do not go through the redcted crossng ont at the same tme. From the vewont of safety and control, the velocty control s more effectve n avodng collson than the drecton control. When usng the velocty control, a decson has to be made as to whch robot should go through the redcted crossng ont frst. In ths aer, two rortes of Fgure 2. Asects under whch two moble robots encounter each other. two moble robots, hgh rorty and low rorty, are frst defned, and the behavors of the two moble robots are then decded accordng to these rortes. As an examle, n Fg. 2Ž. c, the robot wth the hgh rorty s accelerated to go through the redcted crossng ont frst, whle the other robot wth the low rorty s decelerated. In Fg. 2Ž. e, where v2v 1, both the drecton control and the velocty control should be used to avod a collson. For examle, the drecton of robot R2 could be changed and ts velocty ncreased, whle the drecton of robot R1 could be ket constant and ts velocty decreased. These resonses tend to reduce the danger of collson as quckly as ossble. Fgure 2Ž b. s an ntermedate asect between Fgs. 2Ž. a and 2Ž. c, whle Fg. 2Ž. d s between Fgs. 2Ž. c and2ž. e. For these asects, the drecton control and the velocty control are used nterolatvely. The detals wll be gven later. C. Defnton of Prortes of Two Moble Robots As descrbed n the revous secton, a redcted collson s avoded so that the robot wth the hgh
5 Fujmor et al.: Collson Avodance between Robots 351 rorty acts ror to the other robot wth the low rorty. The hgh rorty should be then assgned to a robot that wll arrve at the redcted crossng ont frst. That s, the rortes of two moble robots are defned as the nverse of ther arrval tme at the redcted crossng ont n ths aer. In Fgs. 2Ž b., 2Ž c., and 2Ž d., the redcted crossng ont P s calculated by usng Ž x, y,. Ž 1, Denotng the dstance between the oston of robot R and the redcted crossng ont as RP 12, the rortes w Ž 1, 2. are gven by v w Ž 1,2.. Ž 7. RP 12 If there s no redcted crossng ont because the robots are movng along a straght lne such as shown n Fgs. 2Ž. a and 2Ž. e, the rortes are gven by w v Ž 1,2.. Ž 8. That s, the robot whose velocty s greater s assgned to the hgh rorty. If the rortes are obtaned as w1w 2, robot R1 s assgned to the hgh rorty whle robot R2 s the low rorty n ths aer. where and ch cl are addtonal angles of hgh and low rorty. They are gven by 2 chk sgn 1 Ž 11. cl sgn satž,0,,k,0, 2 / Ž 12. where 1 x0 sgn x ½ Ž x0 satž x, a, b, c, d. c xa dc Ž xa. c axb Ž 14. ba d xb. k s the maxmum of, satž. c s the saturaton functon and s shown n Fg. 3, and and ch cl are as shown n Fg. 4. Smlarly, let v c be the velocty command for the cooeratve avodance mode. It s gven by v ch ww j v, j1 or2, j, Ž 15. c ½ vcl wwj D. Drecton Angle Command and Velocty Command Ths secton resents a desgn of the drecton angle command and the velocty command to realze the cooeratve collson avodance behavors descrbed n Secton III B. Let Ž. c t be the drecton angle command for the cooeratve avodance of robot R. It s gven by where ž / v chsat,0,,v 0, vmax 2 Ž 16. v clsatž,0,,v 0, v mn, 2 / Ž 17., 9 c c where c s an addtonal angle for the cooera- tve avodance mode. As mentoned n Secton III B, the cooeratve collson avodance behavors deend on the rorty and the crossng angle. Accordng to the rorty, s defned as c ½ ch ww j, j1 or2, j, Ž 10. c cl wwj Fgure 3. Functon sat x, a, b, c, d.
6 352 Journal of Robotc Systems 2000 Fgure 4. Ž t.. c Addtonal angle for cooeratve avodance and v and v ch cl are, resectvely, the velocty com- mands of hgh and low rorty and are shown n Fg. 5. At 0 Fg. 2Ž. a, the drecton control s fully used whle the velocty control s not used. At 2 rad Fg. 2Ž. c, on the other hand, only the velocty control s used accordng to the rorty. Durng 02, and v c c are lnearly nter- olated as shown n Fgs. 4 and 5. Ths corresonds to Fg. 2b. E. Extenson of Cooeratve Collson Avodance Ths secton resents an extenson of ths cooeratve collson avodance to stuatons where the number of moble robots s more than two. In stuatons such as ths, t s to be noted that there exst multle relatons between any two moble robots. A moble robot has nr 1 relatons wth the rest of robots. Therefore, the drecton angle command c and the velocty command v c are decded so as to ntegrate the multle relatons. Consder frst the stuaton shown n Fg. 6 where three moble robots are movng toward each other. By evaluatng l and dl dt Ž, j1, 2, 3, Rj Rj Fgure 6. Three moble robots. j., the danger of collson s detected at two onts, P and P, where P Ž P j j s the redcted crossng ont of robots R to R j. Let wj be the rorty of robot R relatve to robot R j. When the danger of collson of robot R wth robot R j s detected, the rorty w s gven by j v w, Ž 18. j RP j where w s set to w0. If Pj s not found, wjv. Moreover, f the danger of collson s not detected, w 0. Equaton Ž 18. j s an extended ex- resson of Eq. Ž. 7. Table 2 shows the rortes for Fg. 6. The avodance behavors of robots R1 and R3 are resectvely decded accordng to the sngle relaton wth resect to robot R 2. On the other hand, robot R2 has two relatons wth resect to robots R1and R 3. It s referable that the behavor of robot R2 s gven by evaluatng all the relatons n whch t s nvolved. Then, the avodance behavor of robot R s determned by takng the center of gravty whose weght s the rorty w. Then, j c Table 2. Fg. 6. Weghts between three moble robots of Fgure 5. v c. Velocty command for cooeratve avodance Oonent R R R R1 0 w12 0 R2 w21 0 w23 R3 0 w32 0
7 Fujmor et al.: Collson Avodance between Robots 353 and v are gven by c Ý n r j1 j cj n Ý r j1 wj w Ž 19. c w v v, Ž 20. c Ý n r j1 j cj n Ý r j1 wj where and v cj cj are the drecton angle command and the velocty command for the cooeratve collson avodance of robot R wth robot R j, resec- tvely. IV. COLLISION AVOIDANCE CONDITION AND NAVIGATION PARAMETERS As mentoned so far, there are several navgaton arameters assocated wth moble robots, avodance behavors, and mode selecton. They are classfed as follows: derve the avodance condtons: ( A3) Let t0 be the tme when the cooeratve avodance mode s selected. The veloctes of two moble robots are v t v t v ( ) A4 The navgaton velocty v s gven by 0 vmax vmn v. Ž () 1 0. Under assumtons Ž A1. Ž A3., a condton for collson avodance of 0 rad s gven as follows. Theorem 1: Consder an asect under whch two moble robots encounter each other wth the redcted crossng angle 0 rad. Then, a collson between the robots s avoded only f the followng relaton s held durng the cooeratve avodance mode: Robot arameter: r, v,, R 0 v Avodance arameter: k, v, v max mn Mode selecton arameter: d, dv The robot arameters are assocated wth moble robots themselves and are gven n advance. The avodance arameters are assocated wth the cooeratve collson avodance behavors and reman to be desgned. The mode selecton arameters are also to be desgned. Ths secton frst dscusses the condtons for collson avodance between two moble robots wth the navgaton arameters. The condtons are used for desgnng the avodance and the mode selecton arameters. For convenence n the followng sectons, d s relaced by a non-d- mensonal arameter l defned as where 1 cosž l 1. A 4 Ž 24. 1A k rr A. Ž 25. Proof: At 0, only the drecton control s used. Fgure 7 shows a stuaton n whch two moble robots aroach each other closest. The trajectores of the robots are symmetrc wth resect to the redcted crossng ont P 12. Durng the cooeratve avodance mode, s ket at 0 rad. From v 0 d l. Ž 21. 2r R A. Collson Avodance Condton Although collson avodance condtons derved n ths secton are for tycal avodance cases whose redcted crossng angle s 0 and 2 rad, they are used as gudelnes for desgnng the navgaton arameters. The followng assumtons are gven to Fgure 7. Cooeratve avodance for 0 rad.
8 354 Journal of Robotc Systems 2000 Eqs. Ž.Ž 9 12., the drecton angle command for the cooeratve collson avodance s gven by k 1,2. 26 c Ž. Ž. Substtutng Eq. 26 nto Eqs. 1 and 2 and ntegratng them, we have t k tt t v k x Ž t. sn Ž t. sn Ž t.4 x Ž t. v0 Ž 28. y Ž t. cos Ž t. cos Ž t.4 y Ž t. 0 0 k 1,2. Wthout loss of generalty, t0 s gven by t00. The startng onts of robots R1 and R2 can be coordnated as Ž. x t, y t, t 0,0,0, Ž. x t, y t, t 2r l,0, R If rr RP 1 12 R2P12 s satsfed durng the cooeratve avodance mode, a collson s avoded. Let t a be the tme when the dstance of robots R1 and R2 s the shortest durng the cooeratve avodance mode and let ta be the tme when the two moble robots collde wth each other wthout changng ther drectons; that s, x Ž t. Ž l 1. 1 a rr t. Ž 31. a v v 0 0 c velocty v0 s ncreased, the rotatonal acceleraton of the drecton control, that s, and k, must be ncreased to kee a secfed A. () 2 2. Under assumtons Ž A1. Ž A4., a condton for collson avodance of 2 rad s gven as follows. Theorem 2: Consder an asect under whch two moble robots encounter each other wth the redcted crossng angle 2 rad. Then, a collson between the robots s avoded only f the followng relaton s held durng the cooeratve avodance mode, where t b r R gž t,., Ž 34. b v v vmax vmn v Ž v t gž t,. t Ž 1e b.. Ž 36. b v b s the tme whch satsfes the followng equaton: v v ' v t 2 r l v t Ž 1e b. 0. Ž 37. R max b Proof: At 2, only the velocty control s used. Fgure 8 shows a stuaton n whch two moble robots R1 and R2 are encountered where s 2 rad. At tt 0, the startng onts of robots R1 and R can be resectvely coordnated as Ž x,0,. v 2 10 Durng the cooeratve avodance mode, the followng relatons are held: Ž. Ž. t t, y t y t, cos t cos t a a 1 a 1 a 1 a 1 a 32 Usng Eqs. Ž 27. and Ž 28., a suffcent condton for r RP y Ž t. cos Ž t. s gven by R a 1 a 5 y Ž t. 1 a v0 1 r 1. Ž 33. R cos Ž t. k ½ cos Ž t. 1 a 1 a Defnng A as Eq. Ž 25., Eq. Ž 33. s wrtten as Eq. Ž 24.. In Eq. Ž 25.,, k, and rr are roortonal to A, whle v s nversely roortonal. If the navgaton 0 Fgure 8. Cooeratve avodance for 2 rad.
9 Fujmor et al.: Collson Avodance between Robots 355 and Ž 0, y, 2. 20, where x 10, y200. The redcted crossng ont P12 s then at the orgn. If x y and v Ž t. v Ž t v 0, the relaton of the rorty of the two robots s gven by w1w 2. For tt Ž 0., the veloctes of the two robots are 0 v t v t v ve 38 1 max v t v t v ve mn The ostons of the robots are gven by H x Ž t. x v Ž. d t v Ž v t x v t 1e. Ž max H t v y Ž t. y v Ž. d v Ž v t y v t 1e.. Ž mn The worst case for collson avodance s that y 20 x 0 and x ' rrl. For ths case, lettng tb be the tme when robot R1 arrves at the redcted crossng ont P, t satsfes Eq. Ž b. If 2rR y Ž t. x Ž t. 2 b 1 b s satsfed durng the cooeratve avodance mode, the collson s avoded. Usng gt, Ž. defned by Eq. Ž 36., we have b v 2r y t x t 2vg t, y x. R 2 b 1 b b v A suffcent condton for Eq. Ž 42. s gven by Eq. Ž 34.. For a fxed r, gt, Ž. R b v should be large enough to satsfy Eq. Ž 34.. gt, Ž. b v s a monotonously ncreas- ng functon because g g 0, 0 for t, 0. Ž 43. b v t b v Moreover, from Eq. Ž 37., l s also monotonously ncreasng wth resect to t b; we have v l 0, for t 0. Ž 44. b t b Therefore, a large v and a large l are referred to hold Eq. Ž 34.. That s, the acceleraton of the velocty should be great and the dstance-swtch to the cooeratve avodance mode d Ž 2r l. should be long. R B. Desgn of Navgaton Parameters Ths secton resents a desgn of the navgaton arameters k, v max, vmn d, and dv by takng nto consderaton the collson avodance condtons: Theorem 1 and 2. An objectve of moble robots n ths aer s to arrve at ther own goals. Although collsons must be avoded, the trajectores of moble robots should be made short from ther startng onts to goals as much as ossble. Equatons Ž 24. and Ž 34. n Theorems 1 and 2 are suffcent condtons for collson avodance. To revent from overactng the avodance behavors, the navgaton arameters are therefore desgned by takng the values near the lower bound whch satsfy Eqs. Ž 24. and Ž 34. n Theorems 1 and 2. A rocedure of desgnng the navgaton arameters s gven as follows. Ste 1. Gve the values of the robot arameters r R,, and v from the erformance of moble robots used. Decde v max, v mn, and dv from the navgaton objectves. d f s desgned so as to satsfy Theorem 1 n Ref. 12. Ste 2. Draw functon gt, Ž. b v and choose the value of tb near the lower bound. Ste 3. Calculate l by usng Eq. Ž 37.. Obtan d usng Eq. Ž 21.. Ste 4. Usng l obtaned n Ste 3, draw func- ton fž A. defned as 1 fž A. cosž l 1. A 4. Ž 45. 1A A ostve fž A. means that Eq. Ž 24. s held. Choose the value of A near the lower bound. Ste 5. Calculate k by usng Eq. Ž 25.. In Theorems 1 and 2, the dervatve-swtch d v was not dscussed. Snce the mode should be swtched to the cooeratve avodance mode for Case 4 n Table 1, a canddate of dv s zero. It should be adjusted by evaluatng avodance erformance. V. NUMERICAL SIMULATION Ths secton resents numercal smulaton examles to demonstrate the effectveness of the ro-
10 356 Journal of Robotc Systems 2000 osed cooeratve avodance behavors and collson avodance condtons for multle moble robots. Frst, two moble robots cases are shown accordng to the desgn rocedure of the navgaton arameters. Usng the navgaton arameters desgned, general cases n whch the number of robots n s 3 and 5 are shown next. Assumtons Ž A1. r Ž A4. were suosed to be held n these smulaton examles. A. Two Moble Robots Ste 1. The robot arameters used n the numercal smulaton were gven as rr 0.3, v0 4.0, 8.488, v Ž 46. Moreover, v, v, and d were gven as max mn v v 8.0, v 0.0, d max mn v From Ref. 12, d f was gven by Ste 2. Fgure 9 shows the lot of functon gt, Ž. wth The lower bound of collb v v son avodance condton Eq. Ž 34. was rrv The lower bound was then gven as tb 0.22 n Fg. 9. Ste 3. Substtutng the obtaned tb nto Eq. Ž 37., the lower bound of l was calculated as 2.8. To hold Eq. Ž 34., the value of l was chosen as l 3.1 and d was then calculated as Fgure 10 shows the trajectores of two moble robots for 2 rad and l 3.1 to hold Eq. Ž 34.. The trajectores were dslayed for a secfed erod k, and the number wrtten n the fgures ndcates the value of k. Fgure 11 shows the tme resonses of the velocty v Ž sold lne. and the command v Ž dashed lne.. Accordng to the dec- son of the rortes, the rorty of robot R1 was greater than that of robot R 2. It s seen from these fgures that a collson was avoded by usng only the velocty control. Robot R1 wth hgh rorty was accelerated whle robot R2 wth low rorty was decelerated after the cooeratve avodance mode was selected. Ste 4. Fgure 12 shows the lot of functon fž A. where l was gven as l 3.1. A ostve fž A. means that the collson avodance condton s satsfed. Postve ranges of fž A. for A 3.7 are meanngless because the robots turn more than 2 Fgure 9. Functon gt, wth b v v
11 Fujmor et al.: Collson Avodance between Robots 357 Fgure 10. Trajectores of two moble robots for 2 rad and l 3.1. durng the cooeratve avodance mode. The value of A was chosen as A Ste 5. k was calculated as rad 40.5 deg. Fgure 13 shows the trajectores of two moble robots for 0 rad where k rad to hold Eq. Ž 24.. Fgure 14 shows the tme resonses of the drecton angle Ž sold lne. and the command Ž dashed lne.. A collson was avoded by usng only the drecton control. To valdate Theorems 1 and 2, the followng smulaton examles are shown. Fgure 15 shows the trajectores of two moble robots wth l 2.4 whch dd not satsfy Eq. Ž 34., whle Fg. 16 shows the trajectores wth l 3.1 and k 0.22 rad 12.6 deg whch dd not satsfy Eq. Ž 24.. In both cases, collsons occurred near the redcted crossng ont. ( ) B. n 2 Moble Robots r Usng the desgned navgaton arameters n the case of two moble robots, numercal smulaton examles of robots nr s 3 and 5 are shown. Fgure 17 shows the trajectores of three moble robots wth Fgure 11. Tme resonses of velocty v Ž sold lne. and command v Ž dashed lne..
12 358 Journal of Robotc Systems 2000 Fgure 12. Functon fž A. wth l 3.1. Fgure 13. Trajectores of two moble robots for 0 rad.
13 Fujmor et al.: Collson Avodance between Robots 359 Fgure 14. Tme resonses of drecton angle Ž sold lne. and command Ž dashed lne.. dfferent startng onts and goals. Fgure 18 shows the trajectores of fve moble robots. Although some robots had multle relatons wth resect to others, ther avodance behavors were decded by ntegratng the multle relatons and all the robots were navgated to ther own goals wthout collsons. Now, we gve comments on the collson avodance condtons, Eqs. Ž 24. and Ž 34., when nr 2. Snce a robot has multle relatons wth the rest of robots, the collson avodance condtons may be volated even f the navgaton arameters were chosen accordng to the rocedure gven n Secton IV B. For examle, f robot R2 avoded robot R1 accordng to the cooeratve avodance behavors but robot R3 was ostoned very close to the area n whch robot R2 was movng, the avodance condtons between robots R2 and R3 may not had held. In the numercal smulaton, ths henomenon occurred n roorton as the number of robots was ncreased. For three moble robots, collsons were avoded n almost all examles, whle collsons were sometmes observed for fve moble robots esecally where the robots are ostoned densely. To mrove ths, the avodance behavors are deter- Fgure 15. Trajectores of two moble robots for 2 rad and l 2.4.
14 360 Journal of Robotc Systems 2000 Fgure 16. Trajectores of two moble robots for 0 rad, l 3.1, and k 0.22 rad. Fgure 17. Trajectores of three moble robots.
15 Fujmor et al.: Collson Avodance between Robots 361 Fgure 18. Trajectores of fve moble robots. mned by takng nto account not only the crossng angle but also the robot densty n a regon. That s, when ncreasng the densty, t may be a way that robots wth lower rorty sto or go back to hold the avodance condtons. Ths ssue wll be dscussed n a future research. VI. NAVIGATION EXPERIMENT USING PIONEER-1 Ths secton shows a navgaton exerment usng two real moble robots named Poneer-1 to verfy the roosed cooeratve collson avodance. Fgure 19 shows a hoto of Poneer-1. The radus s 0.15 m and the weght s 8 kg. Poneer-1 s a small moble robot develoed by Artfcal Intellgence and Grnnell More of Real World Interface Inc. 13 It contans all of basc comonents for robotcs sensng and navgaton n a real-world envronment, ncludng battery ower, drve motors and wheels, oston encoders, and ultrasonc sonar transducers, all managed va an onboard MC68HC11-based mcrorocessor. Poneer-1 communcates the states of robot and the commands wth a clent comuter through a rado modem. 14 To erform the roosed cooeratve collson avodance technque exermentally, two vehcles of Poneer-1 were used as shown n Fg. 20. The states of the robots Žx Ž t., y Ž t., Ž t., v Ž t.. Ž 1, 2. were n common wth the robots through a memo-lnk. A navgaton rogram n whch the navgaton, cooeratve avodance, and fnal modes were ncluded was erformed by each PC comuter. The navgaton arameters used n the navgaton exerment were gven as r 0.15 m, v 1.6 msec, R 0 8 1sec, sec, v 3.2 msec, v 0 msec, max v mn d 1.2 m, d m, d 0.1 m, k 1. Ž 48. f Substtutng these values nto equatons n Theorem 1, we had fž A because of A Smlarly n Theorem 2, we had rrv , tb sec, and gt, Ž. b v Thus, the collson avodance condtons, Eqs. Ž 24. and Ž 34., were satsfed by these values. Fgure 21 shows three exermental results n whch the two robots encounter each other wth the redcted crossng angle 0, 4, and 2 rad. A sequence of crcles means a trajectory of robot and s dslayed by 4 sec. Accordng to the cooeratve avodance behavors, collsons were avoded and the robots were navgated to ther goals safely n the three cases. v
16 362 Journal of Robotc Systems 2000 Fgure 19. Photo of Poneer-1. VII. CONCLUDING REMARKS Ths aer has resented a new collson avodance technque, called cooeratve collson avodance, for multle moble robots. The detecton of the danger of collson between two moble robots was dscussed wth resect to the geometrc asects of ther aths. The drecton control command and the velocty control command for the cooeratve collson avodance were then roosed. The avodance technque was extended to cases n whch the number of moble robots s more than two. Furthermore, the condtons for collson avodance were consdered wth resect to the navgaton arameters and gudelnes of desgnng the navgaton arameters were obtaned. The effectveness of the roosed technque was demonstrated by means of numercal smulaton and navgaton exerments usng two real moble robots named Poneer-1. Fgure 21. Trajectores of two vehcles of Poneer-1. Fgure 20. Navgaton exermental system usng two vehcles of Poneer-1.
17 Fujmor et al.: Collson Avodance between Robots 363 Ths aer gave an nsght nto the navgaton arameters of multle moble robots from the theoretcal onts of vew. The obtaned results are also helful for the learnng technques to choose the ntal values of the arameters. As further subjects to research, the avodance behavors should be mroved to hold the avodance condtons when ncreasng the number of robots. In ths aer, t was suosed that there was no obstacle n the envronment. The roosed technque should be aled to more comlcated envronments n whch statc and movng obstacles are scattered. REFERENCES 1. T. Fukuda, T. Ueyama, A. Saka, A. Ca, T. Sugura, A. Saka, and K. Yamada, Dynamcally reconfgurable robotc systems Ž21st reort, behavor decsons of autonomous moble robots by multrocess., JSME Trans C 61 Ž 1995., S. Hrose, R. Kurazume, and S. Nagata, Cooeratve ostonng system wth multle robots, J Robot Soc Jaan 13 Ž 1995., J. Ohta, Y. Bue, T. Ara, H. Osum, and K. Suyama, Transferrng control by cooeraton of two moble robots, J Robot Soc Jaan 14 Ž 1996., L.E. Parker, ALLIANCE: an archtecture for fault tolerant multrobot cooeraton, IEEE Trans Robot Automat RA-14 Ž 1998., M.J. Matarc, Mnmzng comlexty n controllng a collecton of moble robots, Proc IEEE Int Conf Robotcs Automat, Nce, France, 1992, M.J. Matarc, Desgnng emergent behavors: from local ntegratons to collectve ntellgence, From Anmals to Anmats 2: Int Conf Smulaton of Adatve Behavor, MIT Press, Cambrdge, MA, 1993, R.C. Arkn, Cooeraton wthout communcaton: multagent schema-based robot navgaton, J Robot Syst 9 Ž 1992., K. Sughara and I. Suzuk, Dstrbuted algorthms for formaton of geometrc atterns wth many moble robots, J Robot Syst 13 Ž 996., L. Shan and T. Hasegawa, Sace reasonng from acton observaton for moton lannng of multlerobots: mutual collson avodance n a narrow assage, J Robot Soc Jaan 14 Ž 1996., C.G. Km and M.M. Trved, A neuro-fuzzy controller for moble robot navgaton and multrobot convoyng, IEEE Trans Sys Man Cybernet B SMC-28 Ž 1998., N. Kubota, T. Moroka, F. Kojma, and T. Fukuda, Adatve behavor of moble robot based on sensory network, JSME Trans C 65 Ž 1999., A. Fujmor, P.N. Nkforuk, and M.M. Guta, Adatve navgaton of moble robots wth obstacle avodance, IEEE Trans Robot Automat RA-13 Ž 1997., ActvMeda, Inc., Poneer-1 Oeraton Manual Verson 1, ActvMeda, Inc., Poneer-1 Software Manual Verson 4.1.2, 1996.
Dynamic Optimization. Assignment 1. Sasanka Nagavalli January 29, 2013 Robotics Institute Carnegie Mellon University
Dynamc Optmzaton Assgnment 1 Sasanka Nagavall snagaval@andrew.cmu.edu 16-745 January 29, 213 Robotcs Insttute Carnege Mellon Unversty Table of Contents 1. Problem and Approach... 1 2. Optmzaton wthout
More informationTo: Professor Avitabile Date: February 4, 2003 From: Mechanical Student Subject: Experiment #1 Numerical Methods Using Excel
To: Professor Avtable Date: February 4, 3 From: Mechancal Student Subject:.3 Experment # Numercal Methods Usng Excel Introducton Mcrosoft Excel s a spreadsheet program that can be used for data analyss,
More informationBehavior-Based Autonomous Robot Navigation on Challenging Terrain: A Dual Fuzzy Logic Approach
Behavor-Based Autonomous Robot Navgaton on Challengng Terran: A Dual Fuzzy Logc Approach 1 Kwon Park and 2 Nan Zhang South Dakota School of Mnes and Technology Department of Electrcal and Computer Engneerng
More informationCalculation of the received voltage due to the radiation from multiple co-frequency sources
Rec. ITU-R SM.1271-0 1 RECOMMENDATION ITU-R SM.1271-0 * EFFICIENT SPECTRUM UTILIZATION USING PROBABILISTIC METHODS Rec. ITU-R SM.1271 (1997) The ITU Radocommuncaton Assembly, consderng a) that communcatons
More informationPriority based Dynamic Multiple Robot Path Planning
2nd Internatonal Conference on Autonomous obots and Agents Prorty based Dynamc Multple obot Path Plannng Abstract Taxong Zheng Department of Automaton Chongqng Unversty of Post and Telecommuncaton, Chna
More informationIEE Electronics Letters, vol 34, no 17, August 1998, pp ESTIMATING STARTING POINT OF CONDUCTION OF CMOS GATES
IEE Electroncs Letters, vol 34, no 17, August 1998, pp. 1622-1624. ESTIMATING STARTING POINT OF CONDUCTION OF CMOS GATES A. Chatzgeorgou, S. Nkolads 1 and I. Tsoukalas Computer Scence Department, 1 Department
More informationDetermination of the Multi-slot Transmission in Bluetooth Systems with the Estimation of the Channel Error Probability
Determnaton of the Mult-slot Transmsson n Bluetooth Systems wth the Estmaton of the Channel Error Probablty K Won Sung and Chae Y. Lee Det. of Industral Engneerng, KAIST, 373-1 Kuseong Dong, Taejon, Korea
More informationThe Pilot Alignment Pattern Design in OFDM Systems
The Plot Algnment Pattern Desgn n OFDM Systems *, Yong Chan Lee, Won Chol Jang, 3 Un Kyong Choe, Gyong Chol Leem,3 College of Comuter Scence, Km Il Sung Unversty, D.P.R.K Emal: leeyongchan@yahoo.com Emal:
More informationTEMPERATURE PREDICTION IN TIMBER USING ARTIFICIAL NEURAL NETWORKS
TEMPERATURE PREDICTION IN TIMBER USING ARTIFICIAL NEURAL NETWORKS Paulo Cachm ABSTRACT: Neural networks are a owerful tool used to model roertes and behavour of materals n many areas of cvl engneerng alcatons.
More informationWalsh Function Based Synthesis Method of PWM Pattern for Full-Bridge Inverter
Walsh Functon Based Synthess Method of PWM Pattern for Full-Brdge Inverter Sej Kondo and Krt Choesa Nagaoka Unversty of Technology 63-, Kamtomoka-cho, Nagaoka 9-, JAPAN Fax: +8-58-7-95, Phone: +8-58-7-957
More informationWe assume a two-layered model, and a P wave is excited in the upper layer.
3.4 Crtcal angle We assume a two-layered model, and a P wave s ected n the uer layer. When a P wave mnges on a horontal boundary, the ncdence angle for the transmtted P wave n the second medum s gven by
More informationA MODIFIED DIFFERENTIAL EVOLUTION ALGORITHM IN SPARSE LINEAR ANTENNA ARRAY SYNTHESIS
A MODIFIED DIFFERENTIAL EVOLUTION ALORITHM IN SPARSE LINEAR ANTENNA ARRAY SYNTHESIS Kaml Dmller Department of Electrcal-Electroncs Engneerng rne Amercan Unversty North Cyprus, Mersn TURKEY kdmller@gau.edu.tr
More informationDESIGN AND IMPLEMENTATION OF NETWORKED PREDICTIVE CONTROL SYSTEMS. S C Chai, G P Liu and D Rees
DESIGN AND IMPLEMENTATION OF NETWORKED PREDICTIVE CONTROL SYSTEMS S C Cha, G P Lu and D Rees School of Electroncs, Unversty of Glamorgan, Pontyrdd CF37 DL, UK Abstract: Ths aer dscusses the desgn and mlementaton
More informationControl of Chaos in Positive Output Luo Converter by means of Time Delay Feedback
Control of Chaos n Postve Output Luo Converter by means of Tme Delay Feedback Nagulapat nkran.ped@gmal.com Abstract Faster development n Dc to Dc converter technques are undergong very drastc changes due
More informationKey-Words: - Automatic guided vehicles, Robot navigation, genetic algorithms, potential fields
Autonomous Robot Navgaton usng Genetc Algorthms F. ARAMBULA COSIO, M. A. PADILLA CASTAÑEDA Lab. de Imágenes y Vsón Centro de Instrumentos, UNAM Méxco, D.F., 451 MEXICO Abstract: - In ths paper s presented
More informationLetters. Evolving a Modular Neural Network-Based Behavioral Fusion Using Extended VFF and Environment Classification for Mobile Robot Navigation
IEEE RANSACIONS ON EVOLUIONARY COMPUAION, VOL. 6, NO. 4, AUGUS 2002 413 Letters Evolvng a Modular Neural Network-Based Behavoral Fuson Usng Extended VFF and Envronment Classfcaton for Moble Robot Navgaton
More informationantenna antenna (4.139)
.6.6 The Lmts of Usable Input Levels for LNAs The sgnal voltage level delvered to the nput of an LNA from the antenna may vary n a very wde nterval, from very weak sgnals comparable to the nose level,
More informationInternational Steering Committee:
14 th Internatonal Worksho on ROBOTICS IN ALPE-ADRIA-DANUBE REGION RAAD'05 BUCHAREST ROMANIA : MAY 6-8 005 Internatonal Steerng Commttee: G. Belforte Poltecnco d Torno Italy J. F. Bto Centre of Robotcs
More informationQueen Bee genetic optimization of an heuristic based fuzzy control scheme for a mobile robot 1
Queen Bee genetc optmzaton of an heurstc based fuzzy control scheme for a moble robot 1 Rodrgo A. Carrasco Schmdt Pontfca Unversdad Católca de Chle Abstract Ths work presents both a novel control scheme
More informationApplication of Intelligent Voltage Control System to Korean Power Systems
Applcaton of Intellgent Voltage Control System to Korean Power Systems WonKun Yu a,1 and HeungJae Lee b, *,2 a Department of Power System, Seol Unversty, South Korea. b Department of Power System, Kwangwoon
More informationGAME THEORY AND INTERFERENCE AVOIDANCE IN DECENTRALIZED NETWORKS
GAME TEORY AD ITERFERECE AVOIDACE I DECETRALIZED ETWORS Rekha Menon, (MPRG Vrgna Tech, Blacksburg, VA, USA, rmenon@vt.edu) Dr. Allen B. Macenze, (Electrcal and Comuter Engneerng Deartment, Vrgna Tech)
More informationComparative Analysis of Reuse 1 and 3 in Cellular Network Based On SIR Distribution and Rate
Comparatve Analyss of Reuse and 3 n ular Network Based On IR Dstrbuton and Rate Chandra Thapa M.Tech. II, DEC V College of Engneerng & Technology R.V.. Nagar, Chttoor-5727, A.P. Inda Emal: chandra2thapa@gmal.com
More informationHigh Speed ADC Sampling Transients
Hgh Speed ADC Samplng Transents Doug Stuetzle Hgh speed analog to dgtal converters (ADCs) are, at the analog sgnal nterface, track and hold devces. As such, they nclude samplng capactors and samplng swtches.
More informationAssessment of Reliability and Availability of Series-Parallel Sub- Systems
ISSN -58 (Paer) ISSN 5-5 (Onlne) Vol., No., Assessment of Relablty and Avalablty of Seres-Parallel Sub- Systems Suleman K. Deartment of Mathematcs, Faculty of Scence, Faculty of Scence, Federal Unversty,
More informationRejection of PSK Interference in DS-SS/PSK System Using Adaptive Transversal Filter with Conditional Response Recalculation
SERBIAN JOURNAL OF ELECTRICAL ENGINEERING Vol., No., November 23, 3-9 Rejecton of PSK Interference n DS-SS/PSK System Usng Adaptve Transversal Flter wth Condtonal Response Recalculaton Zorca Nkolć, Bojan
More informationMinimal K-Covering Set Algorithm based on Particle Swarm Optimizer
87 JOURAL OF ETWORKS, VOL. 8, O., DECEMBER 3 Mnmal K-Coverng Set Algorthm based on Partcle Swarm Otmzer Yong Hu Chongqng Water Resources and Electrc Engneerng College, Chongqng, 46, Chna Emal: huyong969@63.com
More informationAutomatic Tuning of Model Predictive Control Using Particle Swarm Optimization
Automatc Tunng of Model Predctve Control Usng Partcle Swarm Otmzaton Rohe Suzuk Fukko Kawa Hdeuk Ito Chkash Nakazawa Yoshkazu Fukuama Etaro Aosh Keo Unverst 3-- Hosh Kohoku-ku Yokohama 3-5 Jaan Fuj Electrc
More informationReliability and Quality Improvement of Robotic Manipulation Systems
Yaser Maddah, Al Maddah Relablty and Qualty Improvement of Robotc Manpulaton Systems Yaser Maddah Department of Mechancal and Manufacturng Engneerng Unversty of Mantoba Wnnpeg, MB R3T 5V6 CANADA Al Maddah
More informationDigital Transmission
Dgtal Transmsson Most modern communcaton systems are dgtal, meanng that the transmtted normaton sgnal carres bts and symbols rather than an analog sgnal. The eect o C/N rato ncrease or decrease on dgtal
More informationFigure 1. DC-DC Boost Converter
EE36L, Power Electroncs, DC-DC Boost Converter Verson Feb. 8, 9 Overvew Boost converters make t possble to effcently convert a DC voltage from a lower level to a hgher level. Theory of Operaton Relaton
More informationLearning Ensembles of Convolutional Neural Networks
Learnng Ensembles of Convolutonal Neural Networks Lran Chen The Unversty of Chcago Faculty Mentor: Greg Shakhnarovch Toyota Technologcal Insttute at Chcago 1 Introducton Convolutonal Neural Networks (CNN)
More informationOptimal Placement of PMU and RTU by Hybrid Genetic Algorithm and Simulated Annealing for Multiarea Power System State Estimation
T. Kerdchuen and W. Ongsakul / GMSARN Internatonal Journal (09) - Optmal Placement of and by Hybrd Genetc Algorthm and Smulated Annealng for Multarea Power System State Estmaton Thawatch Kerdchuen and
More informationChess players fame versus their merit
Ths artcle s ublshed n Aled Economcs Letters htt://www.tandfonlne.com/do/full/0.080/350485.05.0435 Chess layers fame versus ther mert M.V. Smkn and V.P. Roychowdhury Deartment of Electrcal Engneerng, Unversty
More informationMobile Node Positioning in Mobile Ad Hoc Network
Moble Node Postonng n Moble Ad Hoc Networ G.Lu, R.M.Edwards, C.Ladas The Unversty of Sheffeld Abstract: Many Ad Hoc networ rotocols and alcatons assume that the locaton of moble nodes can be obtaned by
More informationReduction of Neural Network Training Time Using an Adaptive Fuzzy Approach in Real Time Applications
Internatonal Journal of Informaton and Electroncs Engneerng, Vol., No. 3, May Reducton of Neural Network Tranng Tme Usng an Adatve Fuzzy Aroach n Real Tme Alcatons Hamdreza Rashdy Kanan and Mahd Yousef
More informationAdaptive Modulation for Multiple Antenna Channels
Adaptve Modulaton for Multple Antenna Channels June Chul Roh and Bhaskar D. Rao Department of Electrcal and Computer Engneerng Unversty of Calforna, San Dego La Jolla, CA 993-7 E-mal: jroh@ece.ucsd.edu,
More informationDesign and Simulation of PD and PID Controller for Hybrid Actuator
3-5 พฤษภาคม 554, โรงแรม Phuet Orch Resort an Sa หาดกะรน จ งหว ดภ เก ต Desgn an Smulaton of PD an PID Controller for Hybr Actuator Boontan Srboonrueng 1 1 Deartment of Inustral Eucaton Faculty of Inustral
More informationAutomatic Level Difficulty Adjustment in Platform Games Using Genetic Algorithm Based Methodology
Automatc Level Dffculty Adjustment n Platform Games Usng Genetc Algorthm Based Methodology Nrach Watcharasatharornong Deartment of Comuter Engneerng Faculty of Engneerng Chulalongkorn Unversty Payatha
More informationCooperative perimeter surveillance with a team of mobile robots under communication constraints
213 IEEE/RSJ Internatonal Conference on Intellgent Robots and Systems (IROS) November 3-7, 213. Toyo, Japan Cooperatve permeter survellance wth a team of moble robots under communcaton constrants J.J.
More informationAn Iterative Parameters Estimation Method for Hyperbolic Frequency Modulated Signals with Colored Noise
An Iteratve Parameters Estmaton Method for Hyerbolc Freuency Modulated Sgnals wth Colored Nose Shua Yao and Shlang Fang Abstract Ths aer resents an teratve method for estmatng the startng freuency and
More informationResource Allocation Optimization for Device-to- Device Communication Underlaying Cellular Networks
Resource Allocaton Optmzaton for Devce-to- Devce Communcaton Underlayng Cellular Networks Bn Wang, L Chen, Xaohang Chen, Xn Zhang, and Dacheng Yang Wreless Theores and Technologes (WT&T) Bejng Unversty
More informationThe Clock-Aided Method for GPS Receiver Positioning in an Urban Environment
Internatonal Journal of Comuter and Electrcal Engneerng, Vol. 3, No. 3, June The Cloc-Aded Method for GPS Recever Postonng n an Urban Envronment Yunlong Teng, Ybng Sh, and Zh Zheng Abstract As the sgnals
More informationMultiple Robots Formation A Multiobjctive Evolution Approach
Avalable onlne at www.scencedrect.com Proceda Engneerng 41 (2012 ) 156 162 Internatonal Symposum on Robotcs and Intellgent Sensors 2012 (IRIS 2012) Multple Robots Formaton A Multobctve Evoluton Approach
More informationStudy on Conflict Detection Method with Downlink Aircraft Parameters
Study on Conflct Detecton Method wth Downlnk Arcraft Parameters Atsush SENOGUCHI, and Yutaka FUKUDA Ar Traffc Management Dvson Electronc Navgaton Research Insttute Tokyo, Jaan senoguch@enr.go.j, fukuda@enr.go.j
More informationA New Type of Weighted DV-Hop Algorithm Based on Correction Factor in WSNs
Journal of Communcatons Vol. 9, No. 9, September 2014 A New Type of Weghted DV-Hop Algorthm Based on Correcton Factor n WSNs Yng Wang, Zhy Fang, and Ln Chen Department of Computer scence and technology,
More informationA High-Sensitivity Oversampling Digital Signal Detection Technique for CMOS Image Sensors Using Non-destructive Intermediate High-Speed Readout Mode
A Hgh-Senstvty Oversamplng Dgtal Sgnal Detecton Technque for CMOS Image Sensors Usng Non-destructve Intermedate Hgh-Speed Readout Mode Shoj Kawahto*, Nobuhro Kawa** and Yoshak Tadokoro** *Research Insttute
More informationMeasuring Cooperative Robotic Systems Using Simulation-Based Virtual Environment
Measurng Cooperatve c Systems Usng Smulaton-Based Vrtual Envronment Xaoln Hu Computer Scence Department Georga State Unversty, Atlanta GA, USA 30303 Bernard P. Zegler Arzona Center for Integratve Modelng
More informationA Fuzzy-based Routing Strategy for Multihop Cognitive Radio Networks
74 Internatonal Journal of Communcaton Networks and Informaton Securty (IJCNIS) Vol. 3, No., Aprl 0 A Fuzzy-based Routng Strategy for Multhop Cogntve Rado Networks Al El Masr, Naceur Malouch and Hcham
More informationFigure 1. DC-DC Boost Converter
EE46, Power Electroncs, DC-DC Boost Converter Verson Oct. 3, 11 Overvew Boost converters make t possble to effcently convert a DC voltage from a lower level to a hgher level. Theory of Operaton Relaton
More informationA Current Differential Line Protection Using a Synchronous Reference Frame Approach
A Current Dfferental Lne rotecton Usng a Synchronous Reference Frame Approach L. Sousa Martns *, Carlos Fortunato *, and V.Fernão res * * Escola Sup. Tecnologa Setúbal / Inst. oltécnco Setúbal, Setúbal,
More informationINITIALIZATION OF ROBOTIC FORMATIONS USING DISCRETE PARTICLE SWARM OPTIMIZATION
24 Internatonal Symposum on on Automaton & Robotcs n n Constructon (ISARC 2007) Constructon Automaton Group, I.I.T. Madras INITIALIZATION OF ROBOTIC FORMATIONS USING DISCRETE PARTICLE SWARM OPTIMIZATION
More informationDegree Distribution Optimization in Raptor Network Coding
2 IEEE Internatonal Symosum on Informaton Theory Proceedngs Degree Dstrbuton Otmzaton n Rator Network Codng Nkolaos Thomos and Pascal Frossard Sgnal Processng Laboratory (LTS4) Swss Federal Insttute of
More informationAnalysis of Integrated Navigation with GPS Source Variability
Analyss of Integrated Navgaton wth GS Source Varablty Mahmoud Efatmaneshn School of Surveyng & SIS, Unversty of New South Wales, Sydney, Australa Yong L School of Surveyng & SIS, Unversty of New South
More informationANNUAL OF NAVIGATION 11/2006
ANNUAL OF NAVIGATION 11/2006 TOMASZ PRACZYK Naval Unversty of Gdyna A FEEDFORWARD LINEAR NEURAL NETWORK WITH HEBBA SELFORGANIZATION IN RADAR IMAGE COMPRESSION ABSTRACT The artcle presents the applcaton
More informationMODELING OF PASSENGER WAITING TIME IN INTERMODAL STATION WITH CONSTRAINED CAPACITY ON INTERCITY TRANSIT
9 MODELING OF PASSENGER WAITING TIME IN INTERMODAL STATION WIT CONSTRAINED CAPACITY ON INTERCITY TRANSIT S.K. Jason CANG Professor Deartment of Cvl Engneerng Natonal Tawan Unversty Tae, 067 Tawan Fax:
More informationTh P5 13 Elastic Envelope Inversion SUMMARY. J.R. Luo* (Xi'an Jiaotong University), R.S. Wu (UC Santa Cruz) & J.H. Gao (Xi'an Jiaotong University)
-4 June 5 IFEMA Madrd h P5 3 Elastc Envelope Inverson J.R. Luo* (X'an Jaotong Unversty), R.S. Wu (UC Santa Cruz) & J.H. Gao (X'an Jaotong Unversty) SUMMARY We developed the elastc envelope nverson method.
More informationParticle Filters. Ioannis Rekleitis
Partcle Flters Ioanns Reklets Bayesan Flter Estmate state x from data Z What s the probablty of the robot beng at x? x could be robot locaton, map nformaton, locatons of targets, etc Z could be sensor
More informationThe Performance Improvement of BASK System for Giga-Bit MODEM Using the Fuzzy System
Int. J. Communcatons, Network and System Scences, 10, 3, 1-5 do:10.36/jcns.10.358 Publshed Onlne May 10 (http://www.scrp.org/journal/jcns/) The Performance Improvement of BASK System for Gga-Bt MODEM Usng
More informationUncertainty in measurements of power and energy on power networks
Uncertanty n measurements of power and energy on power networks E. Manov, N. Kolev Department of Measurement and Instrumentaton, Techncal Unversty Sofa, bul. Klment Ohrdsk No8, bl., 000 Sofa, Bulgara Tel./fax:
More informationImproving Device-to-Device Communication Performance in 5G Networks through Joint Power Control and Resource Allocation
Internatonal Journal of Communcatons htt://www.aras.org/aras/ournals/oc Imrovng Devce-to-Devce Communcaton Performance n 5G Networks through Jont Power Control and Resource Allocaton HWANG-CHENG WANG 1,
More informationSensors for Motion and Position Measurement
Sensors for Moton and Poston Measurement Introducton An ntegrated manufacturng envronment conssts of 5 elements:- - Machne tools - Inspecton devces - Materal handlng devces - Packagng machnes - Area where
More informationCooperative Object Transportation Using Virtual Electric Dipole Field
Internatonal Journal o Mechancal Engneerng and Robotcs Research Vol. 5, No. 1, Januar 2016 Cooeratve Object Transortaton Usng Vrtual Electrc Dole Feld Guho Eoh, Jae Do Jeon, and Beom H. Lee Seoul Natonal
More informationComplexity-Optimized Low-Density Parity-Check Codes for Gallager Decoding Algorithm B
Comlexty-Otmzed Low-Densty Party-Check Codes for Gallager Decodng Algorthm B We Yu, Masoud Ardakan, Benjamn Smth, Frank Kschschang Electrcal and Comuter Engneerng Det. Unversty of Toronto Toronto, Ontaro
More informationRC Filters TEP Related Topics Principle Equipment
RC Flters TEP Related Topcs Hgh-pass, low-pass, Wen-Robnson brdge, parallel-t flters, dfferentatng network, ntegratng network, step response, square wave, transfer functon. Prncple Resstor-Capactor (RC)
More informationFinding Proper Configurations for Modular Robots by Using Genetic Algorithm on Different Terrains
Internatonal Journal of Materals, Mechancs and Manufacturng, Vol. 1, No. 4, November 2013 Fndng Proper Confguratons for Modular Robots by Usng Genetc Algorthm on Dfferent Terrans Sajad Haghzad Kldbary,
More informationGuaranteeing Isochronous Control of Networked Motion Control Systems Using Phase Offset Adjustment
Sensors 2015, 15, 13945-13965; do:10.3390/s150613945 OPEN ACCESS sensors ISSN 1424-8220 www.md.com/journal/sensors Artcle Guaranteeng Isochronous Control of Networked Moton Control Systems Usng Phase Offset
More informationAUTOMATIC HYBRID PARTICLE SWARM OPTIMIZATION RECURSIVE CLUSTERING TECHNIQUE AND ITS APPLICATIONS IN RADIAL BASIS FUNCTION NETWORKS MODELING SYSTEMS
In: Neurocomutng Research Develoments ISBN: 978--6-9- Edtor: Hugo A. Svensson,. - 7 Nova Scence Publshers, Inc. No art of ths dgtal document may be reroduced, stored n a retreval system or transmtted n
More informationFigure.1. Basic model of an impedance source converter JCHPS Special Issue 12: August Page 13
A Hgh Gan DC - DC Converter wth Soft Swtchng and Power actor Correcton for Renewable Energy Applcaton T. Selvakumaran* and. Svachdambaranathan Department of EEE, Sathyabama Unversty, Chenna, Inda. *Correspondng
More informationPower Quality Enhancement Using Energy Storage Devices
Australan Journal of Basc and Aled Scences, 5(6): 779-788, 2011 ISSN 1991-8178 Power Qualty Enhancement Usng Energy Storage Devces S. Sajed, F. Khalfeh, T. Karm, Z. Khalfeh Kharg Branch, Islamc Azad Unversty,
More informationDistributed Topology Control of Dynamic Networks
Dstrbuted Topology Control of Dynamc Networks Mchael M. Zavlanos, Alreza Tahbaz-Saleh, Al Jadbabae and George J. Pappas Abstract In ths paper, we present a dstrbuted control framework for controllng the
More informationPerformance Analysis of Multi User MIMO System with Block-Diagonalization Precoding Scheme
Performance Analyss of Mult User MIMO System wth Block-Dagonalzaton Precodng Scheme Yoon Hyun m and Jn Young m, wanwoon Unversty, Department of Electroncs Convergence Engneerng, Wolgye-Dong, Nowon-Gu,
More informationMulti-sensor optimal information fusion Kalman filter with mobile agents in ring sensor networks
Mult-sensor optmal nformaton fuson Kalman flter wth moble agents n rng sensor networs Behrouz Safarneadan *, Kazem asanpoor ** *Shraz Unversty of echnology, safarnead@sutech.ac.r ** Shraz Unversty of echnology,.hasanpor@gmal.com
More informationA Comparison of Two Equivalent Real Formulations for Complex-Valued Linear Systems Part 2: Results
AMERICAN JOURNAL OF UNDERGRADUATE RESEARCH VOL. 1 NO. () A Comparson of Two Equvalent Real Formulatons for Complex-Valued Lnear Systems Part : Results Abnta Munankarmy and Mchael A. Heroux Department of
More informationUNIT 11 TWO-PERSON ZERO-SUM GAMES WITH SADDLE POINT
UNIT TWO-PERSON ZERO-SUM GAMES WITH SADDLE POINT Structure. Introducton Obectves. Key Terms Used n Game Theory.3 The Maxmn-Mnmax Prncple.4 Summary.5 Solutons/Answers. INTRODUCTION In Game Theory, the word
More informationFast Code Detection Using High Speed Time Delay Neural Networks
Fast Code Detecton Usng Hgh Speed Tme Delay Neural Networks Hazem M. El-Bakry 1 and Nkos Mastoraks 1 Faculty of Computer Scence & Informaton Systems, Mansoura Unversty, Egypt helbakry0@yahoo.com Department
More informationPRACTICAL, COMPUTATION EFFICIENT HIGH-ORDER NEURAL NETWORK FOR ROTATION AND SHIFT INVARIANT PATTERN RECOGNITION. Evgeny Artyomov and Orly Yadid-Pecht
68 Internatonal Journal "Informaton Theores & Applcatons" Vol.11 PRACTICAL, COMPUTATION EFFICIENT HIGH-ORDER NEURAL NETWORK FOR ROTATION AND SHIFT INVARIANT PATTERN RECOGNITION Evgeny Artyomov and Orly
More informationResearch of Dispatching Method in Elevator Group Control System Based on Fuzzy Neural Network. Yufeng Dai a, Yun Du b
2nd Internatonal Conference on Computer Engneerng, Informaton Scence & Applcaton Technology (ICCIA 207) Research of Dspatchng Method n Elevator Group Control System Based on Fuzzy Neural Network Yufeng
More informationDistributed Channel Allocation Algorithm with Power Control
Dstrbuted Channel Allocaton Algorthm wth Power Control Shaoj N Helsnk Unversty of Technology, Insttute of Rado Communcatons, Communcatons Laboratory, Otakaar 5, 0150 Espoo, Fnland. E-mal: n@tltu.hut.f
More informationCooperative localization method for multi-robot based on PF-EKF
Scence n Chna Seres F: Informaton Scences 008 SCIENCE IN CHINA PRESS Sprnger www.scchna.com nfo.scchna.com www.sprngerln.com Cooperatve localzaton method for mult-robot based on PF-EKF WANG Lng, WAN JanWe,
More informationA Neural Network Model that Calculates Dynamic Distance Transform for Path Planning and Exploration in a Changing Environment
Proceedngs of the 3 IEEE Internatonal Conference on Robotcs & Automaton Tape Tawan September 14-19 3 A Neural Network Model that Calculates Dynamc Dstance Transform for Path Plannng and Exploraton n a
More informationFall 2018 #11 Games and Nimbers. A. Game. 0.5 seconds, 64 megabytes
5-95 Fall 08 # Games and Nmbers A. Game 0.5 seconds, 64 megabytes There s a legend n the IT Cty college. A student that faled to answer all questons on the game theory exam s gven one more chance by hs
More informationMobile Sensor Deployment and Coverage Using Multi-Agent-based Collective Formation Schemes
Internatonal Journal of Performablty Engneerng, Vol. 8, No. 2, March, 2012, pp. 141-150. RAMS Consultants Prnted n Inda Moble Sensor Deployment and Coverage Usng Mult-Agent-based Collectve Formaton Schemes
More informationCoordinating the Motions of Multiple Robots with Specified Trajectories
Coordnatng the Motons of Multple Robots wth Specfed Trajectores Srnvas Akella Seth Hutchnson Department of Computer Scence Beckman Insttute Rensselaer Polytechnc Insttute Unversty of Illnos, Urbana-Champagn
More informationAC network state estimation using linear measurement functions
AC network state estmaton usng lnear measurement functons R.A. Jabr and B.C. Pal Abstract: The real/reactve ower and current magntude measurements can be accounted for n an AC network state estmator usng
More informationOptimal Sizing and Allocation of Residential Photovoltaic Panels in a Distribution Network for Ancillary Services Application
Optmal Szng and Allocaton of Resdental Photovoltac Panels n a Dstrbuton Networ for Ancllary Servces Applcaton Reza Ahmad Kordhel, Student Member, IEEE, S. Al Pourmousav, Student Member, IEEE, Jayarshnan
More informationResponse-Time Control of a Single Server Queue
Resonse-Tme Control of a Sngle Server Queue Kjaer, Martn Ansbjerg; Khl, Mara; Robertsson, Anders Publshed n: Proc. of the 46th IEEE Conference on Decson and Control 27 Lnk to ublcaton Ctaton for ublshed
More informationBattery-aware Static Scheduling for Distributed Real-time Embedded Systems
Battery-aware Statc Schedulng for Dstrbuted Real-tme Embedded Systems Jong Luo and Nraj K. Jha Deartment of Electrcal Engneerng Prnceton Unversty, Prnceton, NJ, 08544 {jongluo, jha}@ee.rnceton.edu Abstr
More informationA NOVEL MUSIC BEAT DETECTION ALGORITHM BASED ON PERIODICITY OF ENERGY FLUX. Ajit V. Rao
A NOVEL MUSIC BEAT DETECTION ALGORITHM BASED ON PERIODICITY OF ENERGY FLUX Pradee J. Undergraduate Student Deartment of Electroncs and Communcaton, NITK, Surathkal 575 025. radeejraman@yahoo.com Ajt V.
More informationTest 2. ECON3161, Game Theory. Tuesday, November 6 th
Test 2 ECON36, Game Theory Tuesday, November 6 th Drectons: Answer each queston completely. If you cannot determne the answer, explanng how you would arrve at the answer may earn you some ponts.. (20 ponts)
More informationMulti-Robot Map-Merging-Free Connectivity-Based Positioning and Tethering in Unknown Environments
Mult-Robot Map-Mergng-Free Connectvty-Based Postonng and Tetherng n Unknown Envronments Somchaya Lemhetcharat and Manuela Veloso February 16, 2012 Abstract We consder a set of statc towers out of communcaton
More informationThe Spectrum Sharing in Cognitive Radio Networks Based on Competitive Price Game
8 Y. B. LI, R. YAG, Y. LI, F. YE, THE SPECTRUM SHARIG I COGITIVE RADIO ETWORKS BASED O COMPETITIVE The Spectrum Sharng n Cogntve Rado etworks Based on Compettve Prce Game Y-bng LI, Ru YAG., Yun LI, Fang
More informationThroughput Maximization by Adaptive Threshold Adjustment for AMC Systems
APSIPA ASC 2011 X an Throughput Maxmzaton by Adaptve Threshold Adjustment for AMC Systems We-Shun Lao and Hsuan-Jung Su Graduate Insttute of Communcaton Engneerng Department of Electrcal Engneerng Natonal
More informationKinematics of a dedicated 6DOF Robot for Tele-echography
Knematcs of a dedcated DOF Robot for ele-echography L Al Basst G Posson P Veyres Laboratory of Vson and Robotcs Unversty of Orleans 1800 Bourges France lbasst@bourgesunv-orleansfr Abstract hs paper presents
More informationMTBF PREDICTION REPORT
MTBF PREDICTION REPORT PRODUCT NAME: BLE112-A-V2 Issued date: 01-23-2015 Rev:1.0 Copyrght@2015 Bluegga Technologes. All rghts reserved. 1 MTBF PREDICTION REPORT... 1 PRODUCT NAME: BLE112-A-V2... 1 1.0
More informationLow Switching Frequency Active Harmonic Elimination in Multilevel Converters with Unequal DC Voltages
Low Swtchng Frequency Actve Harmonc Elmnaton n Multlevel Converters wth Unequal DC Voltages Zhong Du,, Leon M. Tolbert, John N. Chasson, Hu L The Unversty of Tennessee Electrcal and Computer Engneerng
More informationWi-Fi Indoor Location Based on RSS Hyper-Planes Method
Chung Hua Journal of Scence and Engneerng, Vol. 5, No. 4, pp. 7-4 (007 W-F Indoor Locaton Based on RSS Hyper-Planes Method Ch-Kuang Hwang and Kun-Feng Cheng Department of Electrcal Engneerng, Chung Hua
More informationNATIONAL RADIO ASTRONOMY OBSERVATORY Green Bank, West Virginia SPECTRAL PROCESSOR MEMO NO. 25. MEMORANDUM February 13, 1985
NATONAL RADO ASTRONOMY OBSERVATORY Green Bank, West Vrgna SPECTRAL PROCESSOR MEMO NO. 25 MEMORANDUM February 13, 1985 To: Spectral Processor Group From: R. Fsher Subj: Some Experments wth an nteger FFT
More informationA MODIFIED DIRECTIONAL FREQUENCY REUSE PLAN BASED ON CHANNEL ALTERNATION AND ROTATION
A MODIFIED DIRECTIONAL FREQUENCY REUSE PLAN BASED ON CHANNEL ALTERNATION AND ROTATION Vncent A. Nguyen Peng-Jun Wan Ophr Freder Computer Scence Department Illnos Insttute of Technology Chcago, Illnos vnguyen@t.edu,
More information(1) i. In this paper it is assumed that the process dynamics can be described with reasonable accuracy by a second-order-plusdead-time
IEEE-TTTC Int Conf on Automaton, Qualty and Testng, Robotcs AQTR, -7, (Eds Mclea and Stoan, ISBN -444-0360-X, Cluj-Naoca, Romana, 006 Develoment and Evaluaton of a PID Auto-Tunng Controller Ioan Naşcu,
More informationN( E) ( ) That is, if the outcomes in sample space S are equally likely, then ( )
Stat 400, secton 2.2 Axoms, Interpretatons and Propertes of Probablty notes by Tm Plachowsk In secton 2., we constructed sample spaces by askng, What could happen? Now, n secton 2.2, we begn askng and
More information