Google Summer of Code/2010/AcceptedProjects/TimeTravelAnalysis

From OpenStreetMap Wiki
Jump to navigation Jump to search

Project Overview

Title

Travel time analysis

People

  • Student: Lukas Kabrt (lukas@kabrt.cz)
  • Mentor: Eric Marsden

Objectives

The goal of this project is to develop an application which can estimate travel times along roads from GPS traces.

The project has two main parts

  • GPS traces to OSM roads matching algorithm (should be able to process avarage quality GPS traces and low-frequency GPS traces)
  • Travel times estimation algorithm (should be able to recognize traffic patters through a week - rush hours)

Wiki Page

Routing/Travel_Time_Analysis

Outline Programme

  • Week 1: Utilities for preprocessing of GPX files and OSM files (OSM4routing)
  • Weeks 2 - 5: Map matching
    • Week 2: Candidates identification, spatial analysis, temporal analysis
    • Week 3: Results matching
    • Week 4: Testing, optimalization
    • Week 5: Attempt to deterime position of car more precisely based on accelaration / deceleration before corners
  • Week 6 - 10: Travel times estimation
    • Week 6: Estimation of free flow time, trafic signals delay
    • Week 7: Estimation of congestion / heavy traffic delay
    • Week 8: Results verification, adjusting algorithm
    • Week 9: Vacation
    • Week 10: Optimalization
  • Week 11: Demonstration on real data
  • Week 12: Reserve, Documentation

Current Outstanding Issues

I need to resolve the following issues to complete the project:

  • find sample GPS traces

Weekly Reports

Community bonding period

Wiki page, SVN repository

Study of scientific articles related to this project, choose algorithms to use

Code for working with GPX files

Week 1

Progress Update for Week Ending 2010-05-30.

Achievements this week:

Wrote a utility that can process OSM files to mapmatching/routing-friendly format. It extracts roads from the OSM file, split them at cross-roads and adds information about accessibility of the road and maximal speed. Details: Routing/Travel_Time_Analysis#OSM2Routing

Wrote a code for GPX preprocessing - it can detect points where the GPS unit wasn't moving

Problems Encountered: nothing serious

Help Required:

Progress Against Plan: according to the plan


Week 2

Progress Update for Week Ending 2010-06-06.

Achievements this week:

Map-matching algorithm:

  • underlaying structures and functions
  • spatial analysis of GPS tracks

Problems Encountered: nothing serious, busy with the last exam

Help Required:

Progress Against Plan: slightly behind the schedule, but I hope I will catch up next week

Week 3

Progress Update for Week Ending 2010-06-13.

Achievements this week:

finished a first working version of the map-matching algorithm

Problems Encountered: algorithm works well with low-frequency gps data, but has some issues with high frequency gps data

Help Required:

Progress Against Plan: back on the schedule


Week 4

Progress Update for Week Ending 2010-06-20.

Achievements this week:

Done some optimalization - map matching is about 5x faster then in the first version

Found solution to the problems with high frequency gps data, but it still needs some work

Problems Encountered: -

Help Required: -

Progress Against Plan: slightly behind the schedule, optimalization and testing took more time then expected


Week 5

Progress Update for Week Ending 2010-06-27.

Achievements this week:

Tried to determine position of car more precisely based on accelaration / deceleration before /after corners. This concept looked promising, but I found out that it does not fit into the STMatching algorithm, so I wasnt able to use it.

Fixed bugs found working with real data.

Wrote code that filters out short u-turns (which are casused by noise in the GPS logs) from the output file.

Problems Encountered: -

Help Required: -

Progress Against Plan: still slightly behind the schedule, I expected to have map matching part finished by this week, but there is still a bug in the output postprocessing (filtering u-turns)

Week 6

Progress Update for Week Ending 2010-07-04.

Achievements this week:

Finished map-matching utility, fixed last bugs (I hope :-) ). Updated binaries, wiki and readme files.

Started working on the analysis of the matched GPS logs.

Problems Encountered: -

Help Required: -

Progress Against Plan: slightly behind

Week 7

Progress Update for Week Ending 2010-07-11.

Achievements this week:

I was working on the analysis of the matched GPS logs - Estimation of the free flow traffic time, traffic signals delay and congestion delay. The basic functionality is working, but it still requires some tuning.

Problems Encountered: -

Help Required:

I am looking for a comprehensive set of GPS track logs from a city or a small part of the country. I have comprehensive data from one bus line, that I am using for testing my utility, but it would be great to test it on some large scale example. If anyone can provide me such data or suggest some source, it would be great help.

Progress Against Plan: on schedule


Week 8

Progress Update for Week Ending 2010-07-18.

Achievements this week:

I was working on the analysis of the matched GPS logs - testing and making adjustemnts to the analysis workflow and analysis parametres.

Problems Encountered: -

Help Required:

Progress Against Plan: on schedule