Traffic Volume Estimation Using Vehicle Trajectory Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current traffic signal optimization solutions are hindered by limited access to time-location data of approaching vehicles and the partial or incomplete data provided by traditional roadside sensors, leading to inefficiencies in traffic management.
Innovation Solution
A method and system utilizing an arrival traffic model with a Gaussian Mixture Model and Expectation Maximization algorithm to predict vehicle arrival patterns and determine optimal traffic signal timing, even with limited data, ensuring continuous probability representation and flexible modeling of vehicle arrivals.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional roadside sensors are used to collect vehicle data, then traffic signal control can be implemented, but the data obtained is partial or incomplete due to device failure, improper installation, or poor weather
Solution Approach 1:
The patent introduces vehicle-mounted sensors as an intermediary data collection mechanism. These sensors on approaching vehicles directly capture time-location data and transmit it to the traffic signal controller, bypassing the unreliable traditional roadside sensors. This intermediary approach ensures more complete and reliable data collection regardless of roadside sensor status or weather conditions.
2Device complexity
If limited access to time-location data of approaching vehicles is accepted, then system simplicity is maintained, but traffic signal optimization becomes difficult
Solution Approach 1:
The patent implements a self-service data collection approach where approaching vehicles autonomously collect and transmit their own time-location data using onboard sensors. Each vehicle serves itself by capturing its position and timing information, then transmitting this data to the traffic signal controller. This eliminates the need for complex roadside infrastructure while enabling effective traffic signal optimization through complete vehicle arrival data.
3Measurement precision
If Gaussian Mixture Model with Expectation Maximization algorithm is applied to predict arrival patterns, then accurate traffic volume estimation is achieved, but computational complexity increases
Solution Approach 1:
The patent applies the Gaussian Mixture Model with Expectation Maximization algorithm as a preliminary action to learn and store optimal arrival rate parameters from historical data. By pre-computing and storing these parameters during off-peak periods or training phases, the system avoids real-time complex calculations during actual traffic control, thus achieving accurate traffic volume estimation without burdening the real-time processing system.
Data Source
AI summary
The systems and methods of the present disclosure provide accurate traffic volume estimates at intersections using data available from a subset of the total vehicles and can effectively resolve the issue with a low penetration rate of connected vehicles and can be used to optimize traffic lights at intersections. The systems and methods of the present disclosure use vehicle trajectory data to calculate arrival times at intersections and, from this data, use a statistical method to estimate the traffic volume. Traffic arriving at the intersection can be modeled by a distribution and then, using an algorithm, the likelihood of the observed data (limited approaching vehicles data) can be maximized. The parameters from the optimized model can be used to estimate the real traffic volume. The systems and methods of the present disclosure use a Gaussian Mixture Model (GMM) distribution to model incoming vehicles.