Congestion forecast device, congestion forecast data and congestion forecast method

Abstract

An acquisition unit (15) acquires a congestion forecast data. The congestion forecast data comprises various forecast congestion degrees generated as results of various congestion degrees of the entering path sections according to each element, each node, and each combination of entering / exiting path sections corresponding to the node. A forecast unit (17) forecasts the congestion degrees of various path sections based on the congestion forecast data. A range limit unit (105) limits an object range namely a congestion forecast range of the path section with congestion degree to be forecasted. The forecast congestion degree acquisition unit (120-145) acquires the real congestion degree on the first moment of the first path section from outside and applies the element suitable for the first moment, the entering path section corresponding to the first path section, and the acquired real congestion degree on the first moment of the first path section on the congestion forecast data, and acquires the forecasted congestion degree of the second path section on the second moment of the second path section from the congestion forecast range of the first path section entering through the first node.

Description

technical field [0001] The invention relates to a congestion prediction device, congestion prediction data and a congestion prediction method. Background technique [0002] Conventionally, Patent Documents 1 and 2 describe techniques for predicting road congestion levels (congestion levels). In Patent Document 1, congestion prediction data for each road at each time point created by statistically processing past traffic information is stored in advance, and based on the congestion prediction information, the future congestion degree of a point far from the current position is predicted. [0003] In addition, in Patent Document 2, a certain amount of traffic is set for a link, the traffic amount is allocated to a branch portion of the road, and the passage in each link is calculated by simulating driving while eliminating a part of the traffic amount in the link. Traffic. [0004] Patent Document 1: Japanese Patent Laid-Open No. 2007-232727 [0005] Patent Document 2: Japa...

AI Technical Summary

Problems solved by technology

For example, if Figure 12 As shown, even if the traffic volume per unit time is constant, in the case of traffic volume above a certain threshold, congestion will be caused due to conditions such as uphill or traffic lights

Therefore, in order to correctly predict congestion by using traffic volume, it is necessary to properly use external parameters to calculate information such as slope and traffic lights, and the accuracy of estimated values ​​is low, and the amount of calculation is also very large

Images