Browse Prior Art Database

The system and methods of the smart traffic based on prediction

IP.com Disclosure Number: IPCOM000236690D
Publication Date: 2014-May-09

Publishing Venue

The IP.com Prior Art Database

Abstract

A autonomous traffic flow prediction system is presented in this disclosure. The core idea is to use video analysis to get historical and real time traffic information, predict next period traffic flow consider captured data of both given cross and effect from adjacent crosses, and automatically trigger new prediction cycle which keeps the system in good performance and catch up the changes that happen in every second in traffic flow.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 31% of the total text.

Page 01 of 17

The system and methods of the smart traffic based on prediction

Traffic congestion is a severe problem in the cities. Specifically, in developing country, the infrastructure construction cannot catch up with vehicle ownership growth. And, in most cities, urban population density is too high.

There is lack of accurate information of the real-time traffic situation. To deploy vast sensors to capture the accurate vehicle count on the road, the cost is very expensive. So, most traffic control just know rough vehicle numbers on the road by experience, not the exact number.

People cannot get accurate prediction of traffic flow. Since of lack of Real-time and historical traffic situation, for example, vehicle count, average speed, it is impossible to make the well performance prediction. Most times, people just predict by human experience.

A autonomous traffic flow prediction system is present in this disclosure. The core idea is to use video analysis to get historical and real time traffic information, predict next period traffic flow consider captured data of both given cross and effect from adjacent crosses, and automatically trigger new prediction cycle which keeps the system in good performance and catch up the changes that happen in every second in traffic flow.

Video analysis can get real time traffic info. It captures traffic condition (vehicle count, speed, direction) in every second. The captured data not only can be used to present real time traffic but also be used to do prediction which would consume the data timely. Without the video analysis, the real time presentation and timely prediction would not be accurate.

A combined predict model is defined that can mine traffic flow pattern for given cross and consider correlations among crosses around it. The prediction accuracy is improve greatly.

Moving average technology is used to trigger the prediction timely and automatically, which ensures autonomous system.

The core advantages are that would help traffic be smart and efficient. Drivers would try to avoid be blocked by real time and prediction information, traffic control would try to allocate resource to avoid potential congestion in advance. It would reduce the congestion observably.

1


Page 02 of 17

Below is the architecture for implementation.

1. Arrows in the diagram mean data flow.

2. Data Cube

A data cube structure is defined to present each

cross in the road network. Generally, the data cube is used in the OLAP. A cube can be considered

2


Page 03 of 17

a generalization of a three-dimensional (or more) spreadsheet. For example, a company might wish to summarize financial data by product, by

time-period, and by city to compare actual and budget expenses. Product, time, city and scenario (actual and budget) are the data's dimensions. Each cell of the cube holds a number that represents some measure of the business, such as sales, profits, expenses, budget and forecast.

In this disclosure, the key dimensions for cube...