Traffic Signal Phase Prediction Using Probe Data Distributions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for determining information about dynamically managed traffic control signals, which have variable phase durations in response to demand, face challenges due to their unpredictability, as they do not rely on fixed cycle times or predictable timing transitions.
Innovation Solution
A method and system that use data indicative of the durations of multiple instances of a traffic control signal phases to determine a distribution of these durations, allowing for the calculation of the probability of the signal being in a specific phase at future times, enabling prediction of phase timings without relying on fixed cycle times.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If fixed infrastructure with fixed cycle times is used, then prediction accuracy is improved, but device complexity and implementation cost increase
Solution Approach 1:
The patent uses probe data from mobile devices to create a virtual model of traffic signal behavior, copying the essential timing information without requiring physical fixed infrastructure. Multiple mobile devices act as distributed sensors, eliminating the need for expensive fixed traffic sensors while maintaining prediction capability
Solution Approach 2:
The patent replaces the mechanical fixed infrastructure system with a software-based probabilistic prediction system. Instead of relying on physical sensors and fixed cycle timing mechanisms, the system uses statistical analysis of probe data to predict signal phases, substituting physical infrastructure with computational methods
2Adaptability or versatility
If dynamically managed traffic control signals are used, then adaptability is improved, but prediction reliability deteriorates
Solution Approach 1:
The patent embraces the dynamic nature of traffic signals by using probabilistic distributions instead of fixed cycle times. The system adapts to variable phase durations by modeling them as statistical distributions, allowing reliable predictions despite the signals' inherent variability and demand-responsive timing
Solution Approach 2:
The system uses probe data from mobile devices to continuously learn and update the probabilistic models of traffic signal behavior. This feedback mechanism allows the system to adapt to changes in signal patterns over time, maintaining prediction reliability even as dynamically managed signals adjust their timing based on real-time traffic conditions
3Quantity of substance
If third party fixed sensor data is used, then data availability is improved, but loss of information increases
Solution Approach 1:
The patent makes mobile devices serve multiple functions: they act as navigation devices for users, communication devices for connectivity, and simultaneously as traffic data sensors for probe data collection. This multi-functionality eliminates the need for dedicated fixed sensor infrastructure while maintaining data availability across the network
Data Source
AI summary
Data indicative of the durations of multiple instances of different phases of a traffic control signal in a given time period is determined. The data is used to obtain data indicative of a distribution of the durations of each phase. The distribution data is used to obtain data indicative of a probability of the traffic control signal having a given phase at one or more future time. The probability data may be used to provide an expected waiting time when arriving at the signal at a future time and/or a speed recommendation for a vehicle approaching the signal.