東京都市大大学院特別講義「ITによって進化する公共交通の最前線」前編：情報技術編

1. H28 IT :

2. (Twitter @niyalist) ◦ ◦ ◦ ◦ 2008-2010 ◦ 2010-2013 ◦ 2013-

3. 1: 1. 2. 3. 4.

4. 2: 1. 2. 3.

5. 3: 1. 2. 3. Twitter LINE SNS

6. http://www.keishicho.metro.tokyo.jp/kotsu/jikoboshi/koreisha/koreijiko.html

7. 2013 CM

8. / ◦ http://giphy.com/gifs/l41lSVMtkWblDT04M

9. ◦ ◦ ◦ http://www.wired.com/2014/03/rising-mass-transit-ridership/

10. IT

11. LRT BRT Uber Lyft 1

12. IT ◦ IT ◦ IT ◦

13. IT ◦ ◦ ◦ ◦

14. NAVYA ARMA ◦ 15 ◦ 5 13 ◦ ◦ ◦ LIDAR ◦ RTK GPS cm ◦

15. ◦ 15 ARMA 2 BestMile ◦ EPFL ◦ ◦

16. EasyMile ◦ 2014 EZ10 ◦ 12 6 ◦ 20km/h 40km/h ◦ 14 ◦ ◦ ◦ :

17. DeNA EasyMile ◦ Robot Shuttle 2016 8 2016 11

18. 2016 2 -3 ◦ ◦

19. IT IT

20. car2go ◦ ◦ 8 30 ◦ ◦ ◦ ※ ◦ ※ 1 2015 10 http://business.nikkeibp.co.jp/atcl/report/15/110879/102600116/

21. 2009 2011 SF 2013 2015 CMU 40 2015 2016 2016

22. Uber UberX ◦ UberPOOL ◦ Uber X ◦ ◦ Lyft Lyft Line

23. ◦ 27 ◦ Airbnb Uber Lyft ◦ ◦ ◦ ◦ ◦

24. Kutsuplus 2013 ◦ 5 10 45 ◦ 9 ◦ 3.5 ◦ AALT

25. http://www.muotoilutarinat.fi/en/project/kutsuplus/

26. 2015 ◦ 2015 : 15 ◦ 2016 45 ◦ 2017 100

27. chariot ◦ ◦ 2014 ◦ ◦ 28 100 2016 Ford Smart Mobility ◦

28. ◦ 31 20 http://www.nakl.t.u-tokyo.ac.jp/odt/index.html

29. IT

30. vison

31. Moovel ◦ ◦ ◦

32. by 2006 ◦ ◦ JR

33. ◦ ◦ ◦ 2011 6 2013 2 ◦ 70

34. 2011 1 ◦ 10 70 100 : 56:44 ◦ 19 ◦ 28 ◦ 53 ◦ 23 ◦ 7 19% 28% 53% PC 28% 携帯電話 31% スマートフォン 36% 専用設置端末 5%

35. iPhone, Android

36. ◦ ◦ ◦ Web ◦ ◦ ◦ ◦ ◦ 1 ◦ 1 ◦ : google maps →

37. AR AR ◦ ◦ ◦ 次発車まで 5:30 鳥取駅行

38. → 便利なシステム見つけたよ。こんな便利なものがあったのか。今度使ってみようかな。

39. Twitter ◦ ◦ URL

40. ◦ 12 14:23 Android

41. 1. 2. 3. 4.

43. →

44. ◦ ◦ ◦

45. …

46. スマートフォンを⽤いた地下鉄位置情報の実現

47. ◦ 2013 3 ◦ ◦

48. GPS GPS is not available on subway because we cannot receive GPS signal. 55 ?????? ???

49. Wi-Fi Wi-Fi Wi-Fi 56 ???

50. : ◦ GPS

51. 1: / ◦ SubwayPS ◦ ◦ ※ Stockx, T., Hecht, B. and Schoning, J.: SubwayPS: To- wards Smartphone Positioning in Underground Public Transportation Systems, Proc. SIGSPATIAL, pp. 93– 102 (2014).

52. 2: → → / Takamasa Higuchi, Hirozumi Yamaguchi, and Teruo Higashino Tracking Motion Context of Railway Passengers by Fusion of Low-Power Sensors in Mobile Devices. In ISWC, 2015

53. 1: ◦ 281 ◦

54. ◦ ◦ ◦ ◦ ◦ ◦ OS

55. SmartKompano ◦ ◦ GPS

56. ◦ ◦ OS GPS Wi-Fi iOS 8.1 ◦ ◦

57. 300m / ◦ 20

58. iPhone 5S 24 ◦ ◦ ◦ ◦

59. Background Activity ◦ Moves

60. ◦ 5 ◦ ◦

61. 2: ◦ ◦

62. 0 5 10 15 20 25 30 35 40 45 50 0 10 20 30 40 50 60 70 80 Altitude[m] Time[sec] iPhone6-A iPhone6-B iPhone6 Plus

63. : , , , " ", , 56(1), pp.260-272, 2015 1 .

64. Time [hPa]

65. 1: / ◦ 10 (13 ) Time

66. Time → →

67. Time / 6 9 ◦

68. 2: /

69. 2: ※ ◦ ◦ 38.6m 10.9m 11.6m 15.2m 11.7m -4.0m -10.9m -3.2m -12.4m 10.3m -2.6m -8.3m -1.8m ※

70. / 80

71. Swift Swift Swift

72. / 4 3 82 Line Stations Time [min.] 25 50 14 30 18 38 21 30

73. 1: / Recall is 85% to 97%, precision is 67% to 92% 83 Line Precision Recall F-measure Asc 0.913 0.973 0.942 Desc 0.864 0.973 0.914 Asc 0.917 0.929 0.916 Desc 0.735 0.857 0.791 Asc 0.788 0.912 0.845 Desc 0.801 0.947 0.866 Asc 0.669 0.952 0.786 Desc 0.775 0.937 0.846 Total 0.808 0.945 0.836

74. 2: Using actual estimated motion state. n is the number of the station used for estimation. Because of motion state errors, accuracy rate is low. 84 Line Accuracy rate n = 2 n = 3 n = 4 n = 5 Asc 0.348 0.515 0.619 0.667 Desc 0.290 0.455 0.571 0.633 Asc 0.333 0.818 0.833 0.778 Desc 0.333 0.424 0.367 0.333 Asc 0.392 0.583 0.644 0.714 Desc 0.373 0.646 0.533 0.619 Asc 0.281 0.352 0.431 0.417 Desc 0.427 0.537 0.510 0.521 Total 0.327 0.498 0.527 0.553

75. 86

76. ACM SIGSPATIAL Satoshi Hyuga, Masaki Ito, Masayuki Iwai, Kaoru Sezaki, "Estimate a User's Location Using Smartphone's Barometer on a Subway", 5th International Workshop on Mobile Entity Localization and Tracking in GPS-less Environments (MELT 2015), Nov. 2015.

77. ◦ 2 ◦ ◦ ◦ / ◦ / ◦

東京都市大 大学院特別講義 「ITによって進化する公共交通の最前線」 前編：情報技術編

Masaki Ito