| 摘要 |
<p>With regards to traffic amount prediction systems that predict the traffic amount on a road from probe data, it is difficult to define the relationship between speed and traffic amount using only the road type, and there is a limit to the accuracy of the traffic amount that is obtained. This traffic amount prediction system estimates road characteristics such as free speed, critical speed, saturation density, and critical density from probe data, groups these road characteristics into clusters that correspond to the probe data, and creates a travel pattern for each cluster on the basis of the speed and acceleration from the probe data that configures each cluster. Next, when probe data is received, if the road associated with the probe data is not affiliated with any of the clusters, the travel pattern generated from the probe data and the travel pattern for each cluster are compared, road characteristics for clusters that are highly similar are acquired, and the acquired road characteristics are used to predict the traffic amount of the road for which the probe data has been collected.</p> |
