Requirements and potential of GPS-based floating car data for traffic management: Stockholm case study

Abstract

The application of GPS probes in traffic management is growing rapidly as the required data collection infrastructure is increasingly in place in urban areas with significant number of mobile sensors moving around covering expansive areas of the road network. The paper presents the development of a laboratory designed to explore GPS and other emerging traffic and traffic-related data for traffic monitoring and control. It also presents results to illustrate the scope of traffic information that can be provided by GPS-based data, using the city of Stockholm as a case study. The preliminary analysis shows that network coverage, especially during peak weekday hours, is adequate. Further investigation is needed to validate the data, and increase its value through fusion with complementary data from other sources.

Highway Traffic State Estimation with Mixed Connected and Conventional Vehicles

“…where µ and H = H T > 0 are the initial conditions of the estimator (18)- (20), which, in the ideal case in which x(k 0 ) is a Gaussian random variable, represent the mean and auto covariance matrix of x(k 0 ), respectively. The Kalman filter (18)-(22) delivers estimates of the inverse percentagesp i ; using (4) and the available data for q a i , ρ a i , we can obtain estimates for all segment (total) flows and densitiesq i ,ρ i as indicated at the output of the Kalman filter in Fig.…”

Highway Traffic State Estimation With Mixed Connected and Conventional Vehicles

“…Since the system equations here are relatively complex, some tuning of the matrices may be necessary for best estimation results. The initial conditions of the estimator (18)- (20) are given byx…”

“…where µ and H = H T > 0 are the initial conditions of the estimator (18)- (20), which, in the ideal case in which x(k 0 ) is a Gaussian random variable, represent the mean and auto covariance matrix of x(k 0 ), respectively.…”

Improving Public Transport Decision Making, Planning and Operations by Using Big Data: Cases from Sweden and the Netherlands

“…Information concerning current traffic conditions could be generated by analyzing traffic data (such as traffic camera counts, loop detectors, plate recognition or floating car data [33]), public transport vehicular data or integrating these two sources. Various machine learning techniques have been applied, due to their capability to utilize large amounts of data, to reveal complex patterns and to address noise in data streams.…”