Cascaded SVM-MLP Classifier for Urban Traffic State Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional traffic state detection methods are time-consuming and costly, often leading to traffic congestion and unsafe driving conditions, as they rely on expensive fixed sensors or probe vehicles, and existing algorithms for predicting traffic density and classifying traffic states are inefficient.
Innovation Solution
A fusion-based system that combines support vector machine (SVM) and multilayer perceptron (MLP) classification algorithms to create a cascaded two-tier classifier for improved traffic state detection, using simulation data to generate training samples and optimize classifier parameters for enhanced accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional fixed detectors are used for traffic state detection, then detection reliability is improved, but system cost increases significantly
Solution Approach 1:
The patent uses probe vehicles as mobile copies of fixed detectors, equipping them with sensors to collect traffic data (occupancy, speed, flow) while moving through the network. This replaces expensive fixed infrastructure with reusable mobile units that can be deployed flexibly across different locations.
Solution Approach 2:
The system leverages existing probe vehicles that are already traveling on the road for their normal purpose, transforming them into data collection devices. These vehicles self-serve the detection function without requiring dedicated detection infrastructure, reducing system cost while maintaining reliability through multiple distributed data sources.
2Ease of operation
If conventional traffic incident detection algorithms are used, then implementation simplicity is improved, but detection speed deteriorates
Solution Approach 1:
The patent transforms the detection approach by changing from traditional occupancy-based parameters to travel time-based parameters. The system calculates average travel times and their variations as probe vehicles traverse road segments, enabling faster detection of traffic incidents through temporal changes in travel patterns rather than relying on complex multi-detector occupancy analysis.
Solution Approach 2:
The system adds a temporal dimension to traffic state detection by analyzing travel time variations across different time intervals. Instead of only spatial analysis from fixed detectors, the patent incorporates time-based metrics (average travel time, travel time difference between intervals) to detect incidents more rapidly and accurately.
3Device complexity
If conventional traffic state classification methods are used, then algorithm simplicity is improved, but detection accuracy deteriorates
Solution Approach 1:
The patent merges multiple classification algorithms (K-Nearest Neighbors, Support Vector Machine, and Neural Networks) into an ensemble system. Each algorithm processes the same traffic parameters (occupancy, speed, flow, travel time) and their results are combined through voting or averaging, achieving higher detection accuracy while maintaining reasonable algorithmic complexity through modular implementation.
Data Source
AI summary
A system and method that facilitates urban traffic state detection based on support vector machine (SVM) and multilayer perceptron (MLP) classifiers is provided. Moreover, the SVM and MLP classifiers are fused into a cascaded two-tier classifier that improves the accuracy of the traffic state classification. To further improve the accuracy, the cascaded two-tier classifier (e.g., MLP-SVM), a single SVM classifier and a single MLP classifier are fused to determine a final decision for a traffic state. In addition, fusion strategies are employed during training and implementation phases to compensate for data acquisition and classification errors caused by noise and/or outliers.