PROBE VEHICLE LANE IDENTIFICATION AT SIGNALIZED INTERSECTION USING SUPPORT VECTOR MACHINE METHOD

Transportation problems in Indonesia, especially urban areas, can be solved by applying Intelligent Systems (ITS), one of which is using a Global Positioning System (GPS) based probe vehicle system. This system growing because smartphones used as instruments for investigators' vehicles. provides information on the location and speed all investigator vehicles every second, making it possible to estimate real time traffic condition road segment passed vehicle. The implementation this city Manado help improve conditions so many motorists carry while driving. However, problem that GPS smartphone has low accuracy 3 15 meters, data cannot solve transportation at micro level such intersections. purpose study was obtain method identifying lanes through from smartphones. steps achieving research goal are 1) building microscopic model software 2) processing algorithm lane identification flow theory machine learning methods case Support Vector Machine, 3) perform validation. results showed individual queue shock wave each could identify also confirms Machine predict position