Road Network Traffic Control Using Directed Graph-LSTM
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing traffic congestion control measures are primarily single-point and ineffective in managing large-scale, spatially and temporally spreading congestion, lacking a comprehensive approach to improve traffic efficiency.
Innovation Solution
A method and system utilizing a directed graph and a long short-term memory neural network to analyze traffic states and implement circle layer and single-point controls based on detector data, predicting congestion spread and proactively managing it through a directed traffic graph convolutional long short-term memory neural network model.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If single-point control measures are used to manage traffic congestion, then the control implementation is simple, but the effectiveness in managing large-scale spatially and temporally spreading congestion is insufficient
Solution Approach 1:
The patent segments the road network into multiple spatial zones (e.g., congestion core area, surrounding areas, distant areas) and applies differentiated control strategies to each segment. This allows the system to handle large-scale spreading congestion by dividing it into manageable zones with specific control measures, thereby improving effectiveness without requiring overly complex centralized control.
Solution Approach 2:
The patent introduces a spatial dimension to traffic control by considering the spatial-temporal characteristics of congestion propagation. Instead of single-point control, the system implements multi-dimensional control across different spatial zones and time periods, capturing the spreading nature of congestion and enabling more effective regional management.
2Reliability
If comprehensive multi-point control measures are implemented to manage large-scale congestion, then the congestion control effectiveness is improved, but the system complexity increases
Solution Approach 1:
The patent applies local quality by implementing differentiated control strategies for different spatial zones based on their specific congestion characteristics. The congestion core area receives one type of control, surrounding areas receive another, and distant areas receive a third type of control. This localized approach improves effectiveness while avoiding the need for uniformly complex control across the entire network.
Solution Approach 2:
The patent uses historical data and modeling to predict future congestion patterns and implements control measures in advance. By analyzing spatial-temporal characteristics and predicting where congestion will spread, the system can take preliminary control actions in anticipated congestion areas before congestion fully develops, improving effectiveness while reducing the need for complex real-time responses.
3Ease of operation
If traditional traffic control methods are used, then the implementation is straightforward, but the ability to predict and prevent congestion spread is limited
Solution Approach 1:
The patent implements preliminary action by using historical traffic data and spatial-temporal modeling to predict future congestion patterns before they occur. The system identifies potential congestion propagation paths and implements control measures in advance to prevent or mitigate congestion spread, thereby reducing congestion duration while maintaining relatively simple operational procedures.
Solution Approach 2:
The patent incorporates feedback mechanisms that continuously monitor traffic conditions and adjust control strategies based on observed congestion patterns. By analyzing real-time data against predicted patterns, the system can refine its predictions and control actions, improving its ability to prevent and manage congestion spread over time while maintaining operational simplicity through automated feedback loops.
Data Source
AI summary
A method and system for active control of road network traffic congestion, and in particular, to the technical field of traffic congestion control includes: constructing a directed graph according to the positions of detectors in a road network; determining a free-flow reachability matrix of the directed graph and a plurality of neighborhood matrices with different orders according to a free-flow vehicle speed between cross-sections where the detectors are located and the directed graph; calculating a convolution operator of the directed graph within a set time period; inputting the convolution operator of the directed graph within the set time period into a long short-term memory neural network model to obtain a traffic state of each cross-section at each moment within a predicted time period; and determining whether a control method for each cross-section is single-point control or circle layer control.