Building a Spatio-Temporal Driving Database

View Statistics Email Alert RSS Feed

Information

Details

Project title

Code/計畫編號

MOST109-2622-E019-003-CC3

Translated Name/計畫中文名

建立具時空訊息的車行資料庫

Project Coordinator/計畫主持人

Chih-Min Chao

Funding Organization/主管機關

National Science and Technology Council

Department/Unit

Department of Computer Science and Engineering

Website

https://www.grb.gov.tw/search/planDetail?id=13443670

Year

2020

Start date/計畫起

01-06-2020

Expected Completion/計畫迄

31-08-2021

Bugetid/研究經費

450千元

ResearchField/研究領域

資訊科學--軟體

傳統保險只採用性別、年齡等簡單資訊做為保費判斷標準，導致駕駛習慣優良之車主負擔的保費可能比有高出險機率的駕駛還高。近年事故率不停上升，增加的理賠金額迫使保險公司提高保費以取得獲利，非優良駕駛的保費提升幅度遠不足以補足虧損，優良駕駛也因此需負擔更重的保費以支應非優良駕駛的理賠費用，顯見傳統保險是不合理的機制。為了改善傳統保險弊病，開發能依據駕駛習慣計費的UBI車險(Usage-based insurance，UBI)非常重要。要提供UBI車險，必須能統計駕駛行為，因此第一個工作就是建立行車數據。本計畫目標為建立一個存放行車資訊的GPS資料統計系統，收集並統計可供後續建立駕駛評估模型的資料集。此系統在接收GPS資料後，將先進行資料清理以排除誤差過大之數據，接著透過地圖匹配將GPS點資料匹配至道路以還原駕駛實際行進的路徑，最後從資料中取出評估所需資訊存放至資料庫中供後續使用。本計畫有三個主要工作項：架設資料庫、編寫地圖匹配程式及編寫資訊提取程式。第一工作項將設計資料庫結構與欄位並建立資料庫，第二個工作項編寫資料清洗及地圖匹配程式，第三個工作項設計並編寫評估資訊的提取程式。 Traditional insurance only uses simple information such as gender and age as the factors of insurance premium calculation, which leads to a chance that drivers having better driving habits have to afford more premiums than those who have higher risks. Due to the continuously increasing car accident rates, insurance companies have to raise premiums, which makes good drivers afford even more premiums to cover poor drivers’ claim expense. Apparently traditional insurance has some irrational mechanism. In this project, we plan to develop the usage-based insurance (UBI), which charges according to drivers’ driving behaviors. We will build a GPS database system to store information for driver’s risk probability analysis. The received GPS data will be checked to remove outliers and then be mapped to roads to rebuild the actual route drivers passes through. Essential data for further driving behavior analysis will also be extracted and store into the database. This project consists of three tasks: building database, implementing map-matching programs and implementing data extraction programs. In first task, we will design the schema of the database. In second task, we will implement data cleaning and map-matching programs. Finally, we will design and implement data extraction programs.

Keyword(s)

GPS
地圖匹配
資料處理
大數據
車聯網
GPS
map matching
data processing
big data
IoV

DSpace CRIS

Building a Spatio-Temporal Driving Database

Details

Description