Traffic Signal Timing Prediction for Reliable Vehicle Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing vehicle control systems rely on inaccurate traffic light signal prediction, particularly in traffic-actuated systems, leading to unreliable vehicle operation and reduced traffic safety.
Innovation Solution
A computer system utilizing quantile regression neural networks (QRNNs) to predict traffic light signal switches with high accuracy, providing reliable timing parameters for vehicle operation by analyzing signal state data and historical patterns.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional SPaT prediction methods are used with default minEndTime and maxEndTime values, then the system is simple to implement, but the prediction accuracy and reliability are insufficient for reliable vehicle control
Solution Approach 1:
The patent introduces Hidden Markov Models as an intermediary computational layer between the traffic light controller and the vehicle. The HMM processes the raw signal state data and generates probabilistic predictions for minEndTime, likelyTime, and maxEndTime, thereby mediating the transformation from simple signal states to reliable timing parameters that enhance vehicle control decisions.
Solution Approach 2:
The patent transforms the prediction approach by changing from providing only a single likelyTime parameter to providing three distinct timing parameters (minEndTime, likelyTime, maxEndTime) with associated confidence values. This parameter expansion allows the system to convey uncertainty and reliability information, enabling vehicles to make more informed control decisions while maintaining compatibility with existing SPaT message formats.
2Measurement precision
If coarse confidence values are provided for likelyTime parameter, then the SPaT message format remains simple, but the information is not helpful for specific vehicle operating instructions
Solution Approach 1:
The patent segments the single likelyTime parameter into three distinct timing parameters (minEndTime, likelyTime, maxEndTime), each representing a different aspect of the signal switch prediction. This segmentation allows the system to provide granular precision information about when the signal will switch, enabling vehicles to derive specific operating instructions such as whether to accelerate, maintain speed, or decelerate based on the confidence intervals provided.
3Reliability
If default values are used for minEndTime and maxEndTime, then the system implementation is straightforward, but the prediction accuracy is less than required for reliable control information
Solution Approach 1:
The patent implements feedback mechanisms where the HMM continuously processes incoming signal state data and updates its predictions of minEndTime, likelyTime, and maxEndTime based on observed patterns and confidence values. This feedback loop allows the system to adapt to actual traffic light behavior, improving prediction accuracy over time while providing vehicles with progressively more reliable control information.
Data Source
AI summary
Described techniques receive signal state data from a controller controlling one or more signal groups of a traffic light system, with a current signal state of respective signal groups, and information about the time interval elapsed since the last state switch of respective signal groups. A feature vector is generated with a first value indicating the elapsed time since the latest switch of the respective signal group to a pass-state, and a second value indicating the elapsed time since the latest switch of the respective signal group to a stop-state. Based on the current feature vector, a quantile regression neural network predicts a low conditional quantile representing the minimum-end-time, a medium conditional quantile representing the likely-time and a high conditional quantile representing the maximum-end-time for respective signal groups. A message which includes the predicted minimum-end-time, likely-time, and maximum-end-time is provided to a receiving device associated with the vehicle.