Real-Time Traffic Signal Control for Modern Roundabouts by Using Particle Swarm Optimization-Based Fuzzy Controller

Gong, Yue‐Jiao; Zhang, Jun · 2014 · OpenAlex-citations

DOI: 10.48550/arxiv.1408.0689

archive: archived pipeline: cataloged verified

Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)

Summary

This paper addresses the challenge of traffic congestion in modern roundabouts, proposing a novel real-time traffic signal control method to improve efficiency where existing infrastructure cannot be expanded. While fixed-time control methods fail to adapt to fluctuating traffic conditions, and existing real-time fuzzy controllers for crossroads (specifically FUZZY-TURN and FUZZY-JUMP) present trade-offs between traffic smoothness and responsiveness, there is a lack of optimized signal control strategies specifically for roundabouts. The authors aim to combine the advantages of both fuzzy approaches to reduce congestion and improve real-time response. The study introduces a two-layer fuzzy controller termed FUZZY-MIX. The outer layer functions as a coarse-grained FUZZY-JUMP controller, computing the urgency degrees of two phase subsets (north-south and east-west traffic) to determine which subset should be active. The inner layer operates as a FUZZY-TURN controller, calculating the extension time for the current phase within the active subset to ensure smooth traffic flow and prevent jams caused by inconsistent phase switching. To optimize the controller’s performance, the authors employ an offline Particle Swarm Optimization (PSO) algorithm to automatically tune the membership functions of the fuzzy logic system, replacing traditional human-defined parameters. This optimized version is designated as FUZZY-MIX-OPT. The proposed controllers were evaluated through simulations of a four-approach roundabout with two circulatory lanes under sixteen different traffic conditions, including both steady and time-varying flows. The performance of FUZZY-MIX and FUZZY-MIX-OPT was compared against four existing methods: Vehicle Actuated (VA), FUZZY-TURN, and FUZZY-JUMP. The results demonstrated that the FUZZY-MIX controller outperformed all previous controllers in terms of traffic efficiency. Furthermore, the application of PSO to optimize the membership functions resulted in the FUZZY-MIX-OPT controller achieving even higher performance than the standard FUZZY-MIX, effectively balancing real-time responsiveness with traffic safety. The significance of this work lies in its three-fold contribution to intelligent transportation systems. First, it addresses a gap in literature by providing a dedicated real-time signal control method for modern roundabouts. Second, it validates the effectiveness of the hybrid FUZZY-MIX architecture, which mitigates the risks of traffic jams associated with FUZZY-JUMP while improving the responsiveness limitations of FUZZY-TURN. Third, it demonstrates the utility of PSO in optimizing fuzzy logic parameters, offering a more efficient and effective technique for traffic signal control that reduces reliance on subjective human preference in system design.

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed.

Summary generated by qwen3.6-27b-prismaquant on 2026-06-26; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• intersection design