CROSS: A Mixture-of-Experts Reinforcement Learning Framework for Generalizable Large-Scale Traffic Signal Control

• URL: http://arxiv.org/abs/2603.24930v1
• Date: Thu, 26 Mar 2026 01:48:15 GMT
• Title: CROSS: A Mixture-of-Experts Reinforcement Learning Framework for Generalizable Large-Scale Traffic Signal Control
• Abstract summary: We propose CROSS, a novel Mixture-of-Experts (MoE)-based decentralized RL framework for generalizable adaptive traffic signal control (ATSC)<n>We first introduce a Predictive Contrastive Clustering (PCC) module that forecasts short-term state transitions to identify latent traffic patterns, followed by clustering and contrastive learning to enhance pattern-level representation.<n>We further design a Scenario-Adaptive MoE module that augments a shared policy with multiple experts, thus enabling adaptive specialization and more flexible scenario-specific strategies.
• Score: 11.628278541732513
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Recent advances in robotics, automation, and artificial intelligence have enabled urban traffic systems to operate with increasing autonomy towards future smart cities, powered in part by the development of adaptive traffic signal control (ATSC), which dynamically optimizes signal phases to mitigate congestion and optimize traffic. However, achieving effective and generalizable large-scale ATSC remains a significant challenge due to the diverse intersection topologies and highly dynamic, complex traffic demand patterns across the network. Existing RL-based methods typically use a single shared policy for all scenarios, whose limited representational capacity makes it difficult to capture diverse traffic dynamics and generalize to unseen environments. To address these challenges, we propose CROSS, a novel Mixture-of-Experts (MoE)-based decentralized RL framework for generalizable ATSC. We first introduce a Predictive Contrastive Clustering (PCC) module that forecasts short-term state transitions to identify latent traffic patterns, followed by clustering and contrastive learning to enhance pattern-level representation. We further design a Scenario-Adaptive MoE module that augments a shared policy with multiple experts, thus enabling adaptive specialization and more flexible scenario-specific strategies. We conduct extensive experiments in the SUMO simulator on both synthetic and real-world traffic datasets. Compared with state-of-the-art baselines, CROSS achieves superior performance and generalization through improved representation of diverse traffic scenarios.

Related papers

• Wireless Traffic Prediction with Large Language Model [54.07581399989292]
  TIDES is a novel framework that captures spatial-temporal correlations for wireless traffic prediction.<n> TIDES achieves efficient adaptation to domain-specific patterns without incurring excessive training overhead.<n>Our results indicate that integrating spatial awareness into LLM-based predictors is the key to unlocking scalable and intelligent network management in future 6G systems.
  arXiv  Detail & Related papers  (2025-12-19T04:47:40Z)
• Multi-residual Mixture of Experts Learning for Cooperative Control in Multi-vehicle Systems [5.5597941107270215]
  We introduce Multi-Residual Mixture of Expert Learning (MRMEL) for Lagrangian traffic control.<n>MRMEL augments a suboptimal nominal AV control policy by learning a residual correction.<n>We validate MRMEL using a case study in cooperative eco-driving at signalized intersections in Atlanta, Dallas Fort Worth, and Salt Lake City.
  arXiv  Detail & Related papers  (2025-07-14T00:17:12Z)
• CoLLMLight: Cooperative Large Language Model Agents for Network-Wide Traffic Signal Control [7.0964925117958515]
  Traffic Signal Control (TSC) plays a critical role in urban traffic management by optimizing traffic flow and mitigating congestion.<n>Existing approaches fail to address the essential need for inter-agent coordination.<n>We propose CoLLMLight, a cooperative LLM agent framework for TSC.
  arXiv  Detail & Related papers  (2025-03-14T15:40:39Z)
• Unicorn: A Universal and Collaborative Reinforcement Learning Approach Towards Generalizable Network-Wide Traffic Signal Control [13.106167353085878]
  Adaptive traffic signal control (ATSC) is crucial in reducing congestion, maximizing throughput, and improving mobility in rapidly growing urban areas.<n>Recent advancements in parameter-sharing multi-agent reinforcement learning (MARL) have greatly enhanced the scalable and adaptive optimization of complex, dynamic flows in large-scale homogeneous networks.<n>We present Unicorn, a universal and collaborative MARL framework designed for efficient and adaptable network-wide ATSC.
  arXiv  Detail & Related papers  (2025-03-14T15:13:42Z)
• TransferLight: Zero-Shot Traffic Signal Control on any Road-Network [0.6274767633959003]
  TransferLight is a novel framework designed for robust generalization across road-networks.<n>Our hierarchical, heterogeneous, and directed graph neural network architecture effectively captures granular traffic dynamics.<n>We develop a single, weight-tied policy that scales zero-shot to any road network without re-training.
  arXiv  Detail & Related papers  (2024-12-12T20:52:12Z)
• LLM-Assisted Light: Leveraging Large Language Model Capabilities for Human-Mimetic Traffic Signal Control in Complex Urban Environments [3.7788636451616697]
  This work introduces an innovative approach that integrates Large Language Models into traffic signal control systems. A hybrid framework that augments LLMs with a suite of perception and decision-making tools is proposed. The findings from our simulations attest to the system's adeptness in adjusting to a multiplicity of traffic environments.
  arXiv  Detail & Related papers  (2024-03-13T08:41:55Z)
• A Holistic Framework Towards Vision-based Traffic Signal Control with Microscopic Simulation [53.39174966020085]
  Traffic signal control (TSC) is crucial for reducing traffic congestion that leads to smoother traffic flow, reduced idling time, and mitigated CO2 emissions. In this study, we explore the computer vision approach for TSC that modulates on-road traffic flows through visual observation. We introduce a holistic traffic simulation framework called TrafficDojo towards vision-based TSC and its benchmarking.
  arXiv  Detail & Related papers  (2024-03-11T16:42:29Z)
• Adaptive Hierarchical SpatioTemporal Network for Traffic Forecasting [70.66710698485745]
  We propose an Adaptive Hierarchical SpatioTemporal Network (AHSTN) to promote traffic forecasting. AHSTN exploits the spatial hierarchy and modeling multi-scale spatial correlations. Experiments on two real-world datasets show that AHSTN achieves better performance over several strong baselines.
  arXiv  Detail & Related papers  (2023-06-15T14:50:27Z)
• End-to-End Intersection Handling using Multi-Agent Deep Reinforcement Learning [63.56464608571663]
  Navigating through intersections is one of the main challenging tasks for an autonomous vehicle. In this work, we focus on the implementation of a system able to navigate through intersections where only traffic signs are provided. We propose a multi-agent system using a continuous, model-free Deep Reinforcement Learning algorithm used to train a neural network for predicting both the acceleration and the steering angle at each time step.
  arXiv  Detail & Related papers  (2021-04-28T07:54:40Z)
• Multi-intersection Traffic Optimisation: A Benchmark Dataset and a Strong Baseline [85.9210953301628]
  Control of traffic signals is fundamental and critical to alleviate traffic congestion in urban areas. Because of the high complexity of modelling the problem, experimental settings of current works are often inconsistent. We propose a novel and strong baseline model based on deep reinforcement learning with the encoder-decoder structure.
  arXiv  Detail & Related papers  (2021-01-24T03:55:39Z)
• MetaVIM: Meta Variationally Intrinsic Motivated Reinforcement Learning for Decentralized Traffic Signal Control [54.162449208797334]
  Traffic signal control aims to coordinate traffic signals across intersections to improve the traffic efficiency of a district or a city. Deep reinforcement learning (RL) has been applied to traffic signal control recently and demonstrated promising performance where each traffic signal is regarded as an agent. We propose a novel Meta Variationally Intrinsic Motivated (MetaVIM) RL method to learn the decentralized policy for each intersection that considers neighbor information in a latent way.
  arXiv  Detail & Related papers  (2021-01-04T03:06:08Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.