Neural Network Traffic Flow Prediction Using Navigation Flow Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies for predicting traffic flow or travel time are inaccurate as they rely solely on past data and real-time conditions, failing to account for actual traffic dynamics.
Innovation Solution
A method and device that utilize massive navigation route data to obtain navigation flow, combined with real-time and past traffic flow and travel time data, to train a neural network for predicting traffic flow or travel time within a prediction time period.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If prediction is based only on past traffic flow and travel time data, then the prediction model is simple, but the prediction accuracy is low
Solution Approach 1:
The patent combines multiple data sources including past traffic flow, past travel time, real-time traffic flow, real-time travel time, and navigation flow into a unified prediction model. This merging of diverse data types allows the model to capture both historical patterns and current dynamics, significantly improving prediction accuracy while managing complexity through integrated processing
Solution Approach 2:
The patent introduces navigation flow data as an additional dimension to the traditional prediction model. By incorporating navigation route data and deriving navigation flow from it, the model adds a new dimensional perspective that reflects actual user navigation behavior, enabling more accurate predictions without oversimplifying the problem
2Reliability
If prediction uses only past and real-time data, then data collection is easier, but prediction reflects actual traffic dynamics poorly
Solution Approach 1:
The patent performs preliminary processing of navigation route data to extract navigation flow information before incorporating it into the prediction model. This preliminary action involves obtaining massive navigation route data from terminals, processing it to derive navigation flow, and preparing it for integration with traffic flow and travel time data, thereby enriching the information base without complicating the overall data collection process
Solution Approach 2:
The patent uses navigation flow as an intermediary variable that bridges the gap between raw navigation route data and traditional traffic parameters. Navigation flow serves as a mediator that translates user navigation behavior into a format compatible with traffic flow and travel time data, enabling the model to capture actual traffic dynamics while maintaining data compatibility
Data Source
AI summary
A method for predicting traffic flow or a travel time, includes: obtaining navigation route data; obtaining, according to the navigation route data, navigation flow of a road within a prediction time period; obtaining real-time traffic flow, a real-time travel time, past traffic flow, and a past travel time of the road; and training a neural network with the real-time travel time, the past travel time, the real-time traffic flow, the past traffic flow, and the navigation flow of the road to obtain predicted traffic flow or a predicted travel time of the road within the prediction time period.