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, Japan a) k-fujita@ist.osaka-u.ac.jp b) hiromori@ist.osaka-u.ac.jp c) h-yamagu@ist.osaka-u.ac.jp d) higashino@ist.osaka-u.ac.jp e) shimojo@cmc.osaka-u.ac.jp Pedestrian Dead Reckoning (PDR) [4], [5], [6] Activity Recognition 1
Laser Range Scanner 30m 270 Twitter WiFi WiFi - 2013 4 26 3 100 The Lab 2014 2 15 2. GPS GPS [2] Dead Reckoning (a) () (b) () (RSS) [7] RSS [1] 2
Wi-Fi 3. 1 UTM-30LX [11] 30m 270 0.25, 40Hz 30mm 50mm Ethernet URG C Library [9] WiFi LAN AP MAC ID MAC WiFi LAN (1) (2) WiFi 1: (3) WiFi 3 4. 4.1 (m x,m y )=(rcos(θ),rsin(θ)) θ,r UTM-30LX θ [0, 270 ] 0.25 t i θ m i (θ, t) δθ θ,θ + δθ r(θ) r(θ + δθ) θ θ + δθ r(θ) r(θ ) 3
3: 2: 2 2 2 i 4.2 () () 30 50mm 2 100mm 4.3 4: 3 4 a a a Ward 4.4 4
5. MAC 5.1 1 t k C k A k k M k : A k C k t C k A k A + t = A t A t 1 M t + : A + t C t + A t = A t A t 1 Mt : A t Ct M t M t = M t + Mt A t 1 = A t 1 A t Mt : A t Ct C t + C t Ct k PA t (x) i PC t (i) (1) x A t 5: 6 M t 1 (x) Mt (if D(PA t (x) = (x),pt C (M t 1(x))) Th ) arg min j D(PA t (x),pt C (j)) (otherwise) D x t 1 t Ct (2) x A + t M + t (x) =argmin j {D(P t A(x),P t C(j)) j C t C t } 6. 6.1 CLEAR-MOT[3] CLEAR-MOT MOTP MOTA 2 MOTP MOTA () 5 5 6 1 6 (MOTP) (MOTA) 6.1.1 6(a) 1 5
(a) MOTP (b) 6: 50mm 6(b) 100mm 40mm-60mm 0mm-40mm 6.1.2 6.2 LAN 5 3 Wi-Fi AP e1,e2,e3 (p1,p2,p3) AP RSS RSS 7 AP RSS 7: AP 1: MOTA Miss False p. Missmatches MOTA 6 3.2 % 0% 0% 96.8 % 4 6.1 % 0% 1.5 % 92.4 % 5 6.5 % 0% 1.5 % 92.0 % 3 7.0 % 0% 3.6% 89.4% 2 10.8 % 0% 9.1% 80.1% 1 15.7 % 0% 11.4% 72.9% 1 5 (MOTA) 5 ID 8 3 1 8(a) 2 2 8(b) 3 8(c) 2 AP RSS AP RSS AP 1 (Pedestrian1, Pedestrian2) RSS AP 9(a) (Phone1, Phone2) RSS AP 9(b) 9(a), 9(b) Pedestrian1 Phone1, Pedestrian2 Phone2 6
(a) 1 RSS 10 10 2 2 (b) 2 10: ( 1) (c) 3 8: 11: ( 2) (a) 12: ( 3) (b) 9: RSS AP 2 1 11 3 1, 2 12 7 AP e2 7
Lab 14 2014 2 13 The Lab WiFi 13: The Lab. Active Lab. 14: RT AP [12] 7. The Lab [13] 132013 4 26 3 100 JR 8 The KDDI - IT - IT 2012 2016 The Lab. [1] Fod, A., Howard, A. and Mataric, M.: A laser-based people tracker, Robotics and Automation, 2002. Proceedings. ICRA 02. IEEE International Conference on, Vol. 3, IEEE, pp. 3024 3029 (2002). [2] Gu, Y., Lo, A. and Niemegeers, I.: A survey of indoor positioning systems for wireless personal networks, Communications Surveys & Tutorials, IEEE, Vol. 11, No. 1, pp. 13 32 (2009). [3] Keni, B. and Rainer, S.: Evaluating multiple object tracking performance: the CLEAR MOT metrics, EURASIP Journal on Image and Video Processing, Vol. 2008 (2008). [4] Kjaergaard, M. B., Wirz, M., Roggen, D. and Tröster, G.: Detecting pedestrian flocks by fusion of multi-modal sensors in mobile phones, Proceedings of the 14th International Conference on Ubiquitous Computing (Ubi- Comp 12), pp. 240 249 (2012). [5] Kloch, K., Lukowicz, P. and Fischer, C.: Collaborative PDR Localisation with Mobile Phones, Proceedings of the 15th Annual International Symposium on Wearable Computers (ISWC 11), pp. 37 40 (2011). [6] Li, F., Zhao, C., Ding, G., Gong, J., Liu, C. and Zhao, F.: A reliable and accurate indoor localization method using phone inertial sensors, Proceedings of the 14th International Conference on Ubiquitous Computing (UbiComp 12), pp. 421 430 (2012). [7] Lim, H., Kung, L.-C., Hou, J. C. and Luo, H.: Zero- Configuration, Robust Indoor Localization: Theory and Experimentation, INFOCOM 2006. 25th IEEE International Conference on Computer Communications. Proceedings, pp. 1 12 (online), DOI: 10.1109/INFO- COM.2006.223 (2006). [8] Teixeira, T., Jung, D. and Savvides, A.: Tasking networked cctv cameras and mobile phones to identify and localize multiple people, Proceedings of the 12th ACM international conference on Ubiquitous computing, ACM, pp. 213 222 (2010). [9] URG Helper project: URG C Library document. [10] Zhao, H. and Shibasaki, R.: A novel system for tracking pedestrians using multiple single-row laser-range scanners, Systems, Man and Cybernetics, Part A: Systems and Humans, IEEE Transactions on, Vol. 35, No. 2, pp. 8
283 291 (2005). [11] UTM-30LX. [12] 64 (2012). [13] KNOWLEDGE CAPITAL. knowledge capital. 9