Project

Transportation Bigdata

As increasing the variety of transportation, users expects the system to provide a wide variety of choice according to their circumstance. Nowadays, though big railway companies could provide real time information of their vehicles, most of public transportation services such as buses and taxies, and ride share vehicles couldn't cooperate each other and provide sufficient information to users. MaaS (Mobility as a Service) is one of the expected solution.

Issues

We collaborate with Kobe Minato Kanko Bus Inc. which operates route buses in Kobe city every day. The company provides us realtime sensor data (eg. GPS, Car speed, Engine speed, Temperature, Humidity, Air pressure, ...) from the real operating buses every 0.5 second. We analyze them and feedback the results to the company for enhancing the operation management efficiency and usability of passengers.

Main topics are as follows. We achieve higher performance of inferring bus operation states and automatic counting of passenger of getting on/off the buses.

  • Analyzing sensor data which is installed in the buses
  • V2X (Vehicle-to-everything) communication technology
  • Developing yet another sensor
  • Developing bigdata computing systems
  • Field study of edge computing

Publication

  • Hayato Nakashima, Ismail Arai, Kazutoshi Fujikawa, “Passenger Counter Based on Random Forest Regressor Using Drive Recorder and Sensors in Buses,” 2019 IEEE International Conference on Pervasive Computing and Communications Workshops (PerCom Workshops)), pp.561--566, Japan, Mar, 2019.
  • Hayato Nakashima, Ismail Arai, Kazutoshi Fujikawa, “Proposal of a Method for Estimating the Number of Passengers with Using Drive Recorder and Sensors Equipped in Buses,” Proceedings of 2018 IEEE International Conference on Big Data, pp.5379--5381, U.S.A, December, 2018. [IEEE Xplore]
  • Hayato Nakashima, Ismail Arai, Kazutoshi Fujikawa, “Counting Passengers from Images of Drive Recorder Inside Buses by Using Background Subtraction and OpenPose,” Internet Conference 2018, Poster No.15, 2018年11月. [PDF]
  • Takuya Yonezawa, Ismail Arai, Toyokazu Akiyama, Kazutoshi Fujikawa, “Random Forest Based Bus Operation States Classification Using Vehicle Sensor Data,” 2018 IEEE International Conference on Pervasive Computing and Communications Workshops (PerCom Workshops), pp.819--824, Greece, March, 2018. [IEEE Xplore]
  • Hayato Nakashima, Ismail Arai, Kazutoshi Fujikawa, “Proposal of Estimating Number of Passengers Using Images of Drive Recorder Inside Buses,” Student Workshop, MobiCASE 2018, Japan, March, 2018. [PDF]
  • Takuya Yonezawa, Ismail Arai, Toyokazu Akiyama, Kazutoshi Fujikawa, “Proposal of classification method of bus operation states using sensor data,” Proceedings of 2017 IEEE International Conference on Big Data, pp.4781--4783, U.S.A, December, 2017. [IEEE Xplore]