LRS WPAN [1] Bluetooth (10m ) LRS LRS [2] LRS LRS Androd LRS UWB [3] Ubsense [4] ActveBat [5] Crcket [6, 7] 10m PDR [8 11] PDR [9, 11, 12] PDR

Size: px

Start display at page:

Download "LRS WPAN [1] Bluetooth (10m ) LRS LRS [2] LRS LRS Androd LRS UWB [3] Ubsense [4] ActveBat [5] Crcket [6, 7] 10m PDR [8 11] PDR [9, 11, 12] PDR"

わんどあざみ
7 years ago
Views:

1 (LRS) LRS LRS LRS LRS Androd LRS Indoor Localzaton utlzng Trackng Scanners and Moton Sensors Takum Takafuj 1 Kazuhsa Fujta 1 Takamasa Hguch 1 Akhto Hromor 1 Hrozum Yamaguch 1 Teruo Hgashno 1 1. GPS RFID 1 Graduate School of Informaton Scence and Technology, Osaka Unversty (LRS) LRS ( LRS 30m 270 ) LRS

2 LRS WPAN [1] Bluetooth (10m ) LRS LRS [2] LRS LRS Androd LRS UWB [3] Ubsense [4] ActveBat [5] Crcket [6, 7] 10m PDR [8 11] PDR [9, 11, 12] PDR [8,10,11] PDR [13, 14]. PDR PDR LRS PDR

3 2.2 LRS cm 1 m LRS LRS [1, 15, 16] LRS [17, 18] LRS 2.3 LRS [19] LRS RFID LRS RFID RFID RFID RFID LRS [1] LRS [20] ( ) LRS [19] RFID LRS [1] [20] [20] LRS

4 PDR W-F LRS ( ) / LRS / LRS LRS LRS LRS (m x, m y ) = (r cos(θ), r sn(θ)) θ r LRS UTM-30LX [21] θ [0, 270 ] 0.25 t LRS θ m (θ, t) LRS LRS δθ θ θ + δθ r(θ) r(θ + δθ) θ θ + δθ LRS ( ) ( ) 30 LRS 50mm 2 100mm LRS LRS LRS LRS 3 LRS a a

5 3 LRS 5 4 ( 4) a Ward LRS 2 ( 2 ) [2] 2 LRS x X (1) wx + (1 w)x 1 f > 0 X = (1) wx 0 f = 0 w 0 < w < 1 1 w = g g

6 8 (4) θ (t, t) 7 ( ) l a max a mn (2) [2] l = k 4 a max a mn + α (2) k α (2) 3.

6 6 8 (4) θ (t, t) 7 ( ) l a max a mn (2) [2] l = k 4 a max a mn + α (2) k α (2) t ω(t) t (< t) t A θ (t, t) (3) t θ (t, t) = ω(t)dt (3) t θ (t, t) ω(t )(t t 1 ) (4) t <t <t t (4) / / Androd (Nexus S) 40[m] 0.6[m/s] ( ) 1.1[m/s] ( ) 1.6[m/s] ( ) 3 10 (2) k α k = α = [m] 0.01[m] 1 5

7 9 t k l k k 1 k θ k l k θ k (2) (3) (1) (2) k 1 k t k = t k t k 1 T {(t k 1, t k ) t k t k 1 > T } u U(t) U(t) (t k 1, t k ) u d (t k 1, t k ) (5) d (t k 1, t k ) = (t t k 1 )/τ s= (t t k )/τ p t sτ p t (s+1)τ (5) (15 ) 100 ( 600 ) [rad] ( 0.6 ) 0.11[rad] ( 6 ) t LRS U(t) LRS τ (UTM-30LX τ = 25 ms) u U(t) LRS u =< p t nτ,..., p t 2τ, p t τ, p t > t nτ u U(t) LRS u U(t) k u U(t) d (t k 1, t k ) 0.5m 1 u 4.3 U(t) 4.2 U (t) U(t) 5 u U (t) ( ) N (µ l, σl 2) (µ l = 0.03[m], σ l = 0.01[m]) 5 l k:k 4 = l k + l k l k 4 N (5µ l, 5σl 2) u (6) 5 l k:k 4

8 lk:k 4 = k k =k 4 p t (t t k )/τ τ p t (t t k 1 )/τ τ (6) lk:k 4 LRS l k:k 4 u U (t) l k:k 4 L dst (l k:k 4 ) (7) L dst 1 (l k:k 4 ) = 10πσ 2 l { exp ((l k:k 4 l k:k 4 ) 5µ l ) 2 10σl 2 (7) lk:k 4 u U (t) u L dst ( (8)) L dst = t k >t nτ Ldst (l k :k 4) k k mn + 1 (8) k k mn u 5 θ k:k N (µ θ, σ 2 θ ) (µ θ = 0.01[rad], σ θ = 0.11[rad]) u (9) 5 θ k:k 4 θ k:k 4 = arg(p t (t t k)/τ τ arg(p t (t t k 4)/τ τ p t (t t k+0.5)/τ τ ) p t (t t k 4+0.5)/τ τ )(9) k 4 k (3.3.4 ) θ k:k 4 ( (10)) { L dr 1 ( θ k:k 4 ) = exp (( θ k:k 4 θ k:k 4 ) µ θ ) 2 } 2πσ 2 θ 2σ 2 θ (10) Θ θ k:k 4 } (10) u U (t) u L dr Θ = 0.11[rad] L dr = t k >t nτ δ( θ k :k 4)L dr ( θ k :k 4) t k >t nτ δ( θ k :k 4) (11) δ( θ k :k 4) θ k :k 4 > Θ 1 0 u L (12) L dst L dr L = L dst L dr (12) U (t) u L Androd (Nexus S) LRS (UTM-30LX [21]) A 5 ( 11) 6 LRS 1 Androd [m/s] ( ),1.1[m/s] ( ), 1.6[m/s] ( ) Androd 3.3 / LRS LRS / LRS Androd LRS

9 11 LRS LRS ( 5 ) 13 LRS ( 12 ) LRS ( 3 ) 14 LRS LRS 1 LRS LRS LRS LRS Androd LRS 1 LRS LRS

10 (a) 1 (b) 2 (c) 3 (d) 4 (e) 5 (f) 1 (g) 2 (h) 3 () 4 (j) 5 (k) 1 (l) 2 (m) 3 (n) 4 (o) 5 13

11 (a) 1 (b) 2 (c) 3 (d) 4 (e) 5 (f) 1 (g) 2 (h) 3 () 4 (j) 5 (k) 1 (l) 2 (m) 3 (n) 4 (o) 5 14

12 W-F [1] Yusuke Wada, Takamasa Hguch, Hrozum Yamaguch, and Teruo Hgashno. Accurate postonng of moble phones n a crowd usng laser range scanners. In Proceedngs of the 9th IEEE Internatonal Conference on Wreless and Moble Computng, Networkng and Comuncaton (WMob 13), [2] H. Wenberg. Usng the adxl202 n pedometer and personal navgaton applcatons. Techncal Report AN-602, Analog Devces, [3] A.F. Molsch, D. Cassol, Cha-Chn Chong, S. Emam, A. Fort, B. Kannan, J. Karedal, J. Kunsch, H.G. Schantz, K. Swak, and M.Z. Wn. A comprehensve standardzed model for ultrawdeband propagaton channels. IEEE Transactons on Antennas and Propagaton, Vol. 54, No. 11, pp , [4] Ubsense. [5] A. Harter, A. Hopper, P. Steggles, A. Ward, and P. Webster. The anatomy of a context-aware applcaton. Wreless Networks, Vol. 8, No. 2, pp , [6] N.B. Pryantha, A. Chakraborty, and H. Balakrshnan. The Crcket locaton-support system. In Proceedngs of the 6th Annual Internatonal Conference on Moble Computng and Networkng (MobCom 00), pp , [7] M. Hazas and A. Hopper. Broadband ultrasonc locaton systems for mproved ndoor postonng. IEEE Transactons on Moble Computng, Vol. 5, No. 5, pp , [8] B. Krach and P. Robertson. Integraton of footmounted nertal sensors nto a bayesan locaton estmaton framework. In Proceedngs of the 5th Internatonal Workshop on Postonng, Navgaton and Communcaton (WPNC 08), pp , [9] Ulrch Stenhoff and Bernt Schele. Dead reckonng from the pocket an expermental study. In Proceedngs of the 8th Internatonal Conference on Pervasve Computng and Communcatons (PerCom 10), pp , [10] Olver Woodman and Robert Harle. Pedestran localsaton for ndoor envronments. In Proceedngs of the 10th ACM Internatonal Conference on Ubqutous Computng (UbComp 08), pp , [11] Fan L, Chunshu Zhao, Guanzhong Dng, Jan Gong, Chenxng Lu, and Feng Zhao. A relable and accurate ndoor localzaton method usng phone nertal sensors. In Proceedngs of the 14th ACM Internatonal Conference on Ubqutous Computng (UbComp 12), pp , [12] Ionut Constandache, Romt Roy Choudhury, and Injong Rhee. Towards moble phone localzaton wthout war-drvng. In Proceedngs of the 29th Internatonal Conference on Computer Communcatons (IN- FOCOM 10), pp. 1 9, [13] Takamasa Hguch, Hrozum Yamaguch, and Teruo Hgashno. Clearng a crowd: Cotext-supported neghbor postonng for people-centrc navgaton. In Proceedngs of the 10th Internatonal Conference on Pervasve Computng (Pervasve 12), pp , [14] K. Kloch, P. Lukowcz, and C. Fscher. Collaboratve PDR localsaton wth moble phones. In Proceedngs of the 15th Internatonal Symposum on Wearable Computers (ISWC 11), pp , [15] Ajo Fod, Andrew Howard, and Maja J Matarc. Laserbased people trackng. In Proceedngs of the IEEE Internatonal Conference on Robotcs and Automaton (ICRA 02), pp , [16],,,,.. 24, [17],,,,,.., Vol. 88, No. 7, pp , [18] H. Zhao and R. Shbasak. A novel system for trackng pedestrans usng multple sngle-row laser-range scanners. IEEE Transactons on Systems, Man, and Cybernetcs, Part A, Vol. 35, No. 2, pp , [19] Drk Schulz, Deter Fox, and Jeffrey Hghtower. People trackng wth anonymous and d-sensors usng raoblackwellsed partcle flters. In Proceedngs of the 18th Internatonal Jont Conference on Artfcal Intellgence (IJCAI 03), pp , [20] Thago Texera, Deokwoo Jung, and Andreas Savvdes. Taskng networked CCTV cameras and moble phones to dentfy and localze multple people. In Proceedngs of the 12th ACM Internatonal Conference on Ubqutous Computng (Ubcomp 10), pp , [21] Hokuyo Automatc Co., LTD. Scannng range fnder, UTM-30LX.

IPSJ SIG Technical Report Vol.2014-MBL-70 No.49 Vol.2014-UBI-41 No /3/15 2,a) 2,b) 2,c) 2,d),e) WiFi WiFi WiFi 1. SNS GPS Twitter Facebook Twit

IPSJ SIG Technical Report Vol.2014-MBL-70 No.49 Vol.2014-UBI-41 No /3/15 2,a) 2,b) 2,c) 2,d),e) WiFi WiFi WiFi 1. SNS GPS Twitter Facebook Twit 2,a) 2,b) 2,c) 2,d),e) WiFi WiFi WiFi 1. SNS GPS Twitter Facebook Twitter Ustream 1 Graduate School of Information Science and Technology, Osaka University, Japan 2 Cybermedia Center, Osaka University,