Agent-Based Traffic Simulation on GPUs for Real-Time Operation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current state-of-the-art traffic simulation techniques face computational complexity challenges when simulating large-scale transportation systems, making real-time simulations unattainable due to the computational resources required for microscopic-level modeling of numerous agents, such as vehicles and pedestrians.
Innovation Solution
Utilizing graphical processing units (GPUs) to parallelize the calculation of agent-based models, shifting the computational burden from CPUs to GPUs, allowing for efficient parallel processing of agent objects and enabling real-time traffic simulations by distributing the simulation workload across multiple processing units.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If CPU-based microscopic traffic simulation is used to model individual vehicles and pedestrians, then simulation accuracy and behavioral detail are improved, but computational complexity and processing time increase significantly
Solution Approach 1:
The patent replaces CPU-based sequential processing with GPU-based parallel processing architecture. The simulation system utilizes graphics processing units originally designed for graphical computations to perform mass parallel calculations of agent behaviors, substituting the traditional CPU computational mechanism with a more suitable parallel computing architecture that can handle the computational intensity of microscopic traffic simulation.
Solution Approach 2:
The patent segments the traffic simulation into discrete agent objects (vehicles, pedestrians) that can be independently processed. Each agent's behavior calculation is divided into separate computational units that can be executed in parallel across multiple GPU cores, allowing the complex simulation to be broken down into manageable, concurrently executable segments.
2Quantity of substance
If the population of agents is increased to represent large-scale transportation systems, then simulation realism and coverage are improved, but computational tractability deteriorates
Solution Approach 1:
The patent transitions from sequential processing in one dimension (CPU time) to parallel processing across multiple dimensions (GPU cores). By utilizing the spatial parallelism of GPU architecture with thousands of cores, the system can simulate a larger number of agents simultaneously, effectively adding a dimensional aspect to computational processing that enables scaling to large agent populations.
Solution Approach 2:
The patent creates multiple copies of the same simulation logic across numerous GPU cores. Each core executes identical or similar agent behavior calculations for different agents, allowing the system to process large numbers of agents by replicating the computational template across the GPU's parallel processing units rather than sequentially executing the same logic.
3Loss of time
If real-time processing is required for smart traffic services, then service responsiveness is improved, but computational resource requirements worsen
Solution Approach 1:
The patent substitutes traditional CPU-based real-time processing with GPU-based parallel processing to achieve real-time simulation performance. The massive parallel computing power of GPUs enables the system to process large numbers of agents and generate real-time traffic flow information within the required time constraints, overcoming the computational resource limitations of CPU-based systems.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A computer-implemented method for traffic simulation, the method comprising: accepting input of traffic data comprising data for a geographic region; generating a plurality of agent objects using the traffic data, each agent object representing a traffic participant and associated data; allocating the plurality of agent objects to a respective plurality of processing units; for each processing unit in parallel, executing a traffic simulation model for simulating the behaviour of the traffic participant, wherein the traffic simulation model comprises a car-following model, and the traffic simulation model comprises a lane-changing model and/or a junction management model.