Agent-Based Traffic Simulation Using Privacy-Preserving Synthetic Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing traffic simulation models using synthetic data lack accuracy in representing real-world driving behaviors, and using real-world data raises privacy concerns.
Innovation Solution
A system that translates real-world agent-based position data into narrativized instructions, using a machine-learning model to generate synthetic movement data that mimics real-world driving behaviors, while maintaining privacy through synthetic data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If real-world agent-based position data is used for traffic simulation, then accuracy in representing real driving behaviors is improved, but privacy violations occur
Solution Approach 1:
The patent creates synthetic agent-based position data that copies the statistical properties and movement patterns of real driving data without containing actual personal information. The synthetic data is generated through a transformation process that preserves the essential characteristics of real-world traffic behavior while eliminating identifiable individual trajectories, thus achieving accuracy without privacy violations.
Solution Approach 2:
The patent introduces an intermediary transformation process that converts real-world position data into synthetic data. This intermediary step acts as a mediator between the real data (which contains privacy-sensitive information) and the simulation requirements (which need accurate behavioral representation). The transformation maintains the statistical properties needed for accurate simulation while removing direct references to actual individuals.
2Object-affected harmful factors
If synthetic data is used for traffic simulation, then privacy is protected, but accuracy in representing real driving behaviors deteriorates
Solution Approach 1:
The patent applies parameter changes by transforming the data through a mathematical transformation that preserves statistical properties such as distribution patterns, movement speeds, and spatial relationships. The transformation changes the data representation parameters while maintaining the essential characteristics needed for accurate traffic behavior simulation, thus achieving both privacy protection and accuracy.
3Reliability
If real data is used for traffic simulation, then accurate representation of driving behaviors is achieved, but privacy concerns arise
Solution Approach 1:
The patent creates synthetic agent-based position data that copies the statistical properties and movement patterns of real driving data without containing actual personal information. The synthetic data is generated through a transformation process that preserves the essential characteristics of real-world traffic behavior while eliminating identifiable individual trajectories, thus achieving accuracy without privacy violations.
Data Source
AI summary
Implementations claimed and described herein provide systems and methods for simulating traffic using synthetic data based on agent-based modeling that simulates real drivers in a particular geographic area and time frame. In one implementation, inputting, in a machine-learning model of a simulation system, real-world agent-based position data associated with a custom selection of a geographic area and a time frame. The machine-learning model of the simulation system outputs metrics associated with synthetic agent-based position data over time within a map for the simulated environment and the variation of the simulated environment, wherein the metrics represents synthetic movement behavior of agents associated with synthetic individuals based on real movement behavior associated with the geographic area and the time frame.