Preemptive Corridor Clearing Protocol Using RH-MCTS for Emergency Vehicles

Abstract

This work proposes Dynamic Corridor Clearing Protocol (DCCP), which utilizes Rolling Horizon Monte Carlo Tree Search (RH-MCTS) to achieve preemptive signal priority control for EVs in an urban road network. Preemptive implies the act of giving priority in advance before the arrival of the Emergency Vehicle, whereas Corridor clearing involves managing traffic flow at multiple intersections. RH-MCTS is defined as a real-time decision-making procedure involving simulations of possible signal plans within a finite period. The proposed model considers the entire traffic states of a corridor and evaluates the sequence of signal phases through a multi-objective reward system. In this paper, we propose a real-time adaptive solution that aims to optimize EV passage and ensure traffic stability. The model incorporates rolling horizon optimization to iteratively refine the decision-making process based on the current traffic state without any need for pre-training procedures. Simulation studies show improved EV efficiency, controlled queues, and increased throughput compared to the fixed-time strategy.

Introduction

This work addresses the problem of efficient emergency vehicle (EV) navigation through urban traffic signals, where existing systems like fixed-time control, hardware-based preemption, and conventional adaptive traffic systems fail to coordinate multiple intersections effectively. These approaches either improve isolated intersections or optimize general traffic flow, but do not provide a unified, predictive corridor-wide solution for emergency vehicle prioritization.

To overcome these limitations, the study proposes the Dynamic Corridor Clearing Protocol (DCCP), a multi-intersection signal control system based on Rolling Horizon Monte Carlo Tree Search (RH-MCTS). Unlike reinforcement learning methods that require heavy training and lack interpretability, DCCP performs real-time planning without pre-training, exploring possible future signal states and selecting optimal actions using a reward-based search strategy.

The system models traffic corridors as a sequential decision-making problem, where each intersection is controlled by a finite state machine (FSM) enforcing safe signal transitions. Traffic flow is simulated using queue models and mathematical formulations, while emergency vehicle movement is modeled using realistic congestion-based speed functions.

The architecture consists of a Python/FastAPI backend, a simulation engine, an MCTS solver, and React-based dashboards for administrators and drivers. A priority-based event-driven simulation engine tracks signal states, vehicle queues, and emergency vehicle progress in real time.

At the core of the system, the RH-MCTS algorithm evaluates multiple possible signal control actions (e.g., hold, extend, or preempt phases) across intersections within a planning horizon. It uses a structured rollout policy and a multi-objective reward function that prioritizes EV speed while balancing overall traffic congestion.

Key advantages of the proposed system include:

• Predictive multi-intersection corridor clearing
• Real-time adaptive signal control
• Fully interpretable decision-making via search trees
• No requirement for offline training
• Balanced optimization between EV priority and general traffic flow

Experimental design shows that the system can efficiently coordinate signal timing to reduce emergency vehicle delay while preventing excessive disruption to cross traffic. The approach improves over hardware-based preemption and reinforcement learning methods by offering a scalable, explainable, and deployment-friendly solution.

Conclusion

DCCP stands as an effective intermediate strategy bridging the large-grained nature of hardware preemption and the implementation issues surrounding deep reinforcement learning. Considering the multi-intersection signal coordination problem as an online tree search leads to a control algorithm that, while sufficiently adaptable to beat out fixed-timing approaches, is also easily auditable by engineering analysis and computationally efficient enough for execution on edge devices – none of which can be said of the established competitors. Practically speaking, DCCP encompasses qualities lacking in any current systems: a prediction horizon spanning the entire corridor for a period of 60 seconds, traffic queue penalties encoded as an essential part of the reward function, absolute physical safety provided by invariant finite-state machine constraints, and zero extra overhead beyond connectivity. Such qualities make DCCP well-suited as a supervisor priority layer for integrating electric vehicles, especially in the context of Smart Cities with limited budgets and capabilities. An all-encompassing implementation of the system, including backend, two frontends, test harness, JSON configs, and comparators included, forms a runnable reference point for further study and possible implementation by municipalities.

References

[1] S. Shaikh and S. Deokar, “RFID-based emergency vehicle preemption system for urban intersections,” in Proc. Int. Conf. Intelligent Computing and Information Technology (ICITE), Pune, India, 2019, pp. 245–251. [2] G. Nokhbe Zarei, M. Jalili Ghazizadeh, and S. M. Mortazavi, “Evaluation of emergency vehicle preemption strategies on arterial corridors,” Journal of Transportation Engineering, vol. 143, no. 7, pp. 04017021, 2017. [3] Y.-Y. Chen, J.-Y. Wang, S.-C. Lo, and W.-T. Sung, “An emergency vehicle traffic signal preemption system considering queue spillbacks along routes and negative impacts on non-priority traffic,” IET Intelligent Transport Systems, vol. 18, no. 8, pp. 1385–1395, Aug. 2024. [4] B. P. Gokulan and D. Srinivasan, “Distributed geometric fuzzy multiagent urban traffic signal control,” IEEE Trans. Intelligent Transportation Systems, vol. 11, no. 3, pp. 714–727, Sep. 2010. [5] H. Wei, G. Zheng, H. Yao, and Z. Li, “IntelliLight: A reinforcement learning approach for intelligent traffic light control,” in Proc. 24th ACM SIGKDD Int. Conf. Knowledge Discovery and Data Mining, London, UK, 2018, pp. 2496–2505. [6] G. Zheng, X. Xiong, H. Zang, J. Feng, H. Wei, H. Yao, Y. Cong, and Z. Li, “Learning phase competition for traffic signal control,” in Proc. 28th ACM Int. Conf. Information and Knowledge Management (CIKM), Beijing, China, 2019, pp. 1963–1972. [7] W. Chen, S. Yang, W. Li, Y. Hu, X. Liu, and Y. Gao, “Learning multi-intersection traffic signal control via coevolutionary multi-agent reinforcement learning,” IEEE Trans. Intelligent Transportation Systems, vol. 25, no. 11, pp. 15947–15963, Nov. 2024. [8] H. Luo, Y. Bie, and S. Jin, “Reinforcement learning for traffic signal control in hybrid action space,” IEEE Trans. Intelligent Transportation Systems, vol. 25, no. 6, pp. 5225–5241, Jun. 2024. [9] D. Silver, A. Huang, C. J. Maddison, A. Guez, L. Sifre, G. van den Driessche, J. Schrittwieser, I. Antonoglou, V. Panneershelvam, M. Lanctot, S. Dieleman, D. Grewe, J. Nham, N. Kalchbrenner, I. Sutskever, T. Lillicrap, M. Leach, K. Kavukcuoglu, T. Graepel, and D. Hassabis, “Mastering the game of Go with deep neural networks and tree search,” Nature, vol. 529, no. 7587, pp. 484–489, Jan. 2016. [10] C. B. Browne, E. Powley, D. Whitehouse, S. M. Lucas, P. I. Cowling, P. Rohlfshagen, S. Tavener, D. Perez, S. Samothrakis, and S. Colton, “A survey of Monte Carlo tree search methods,” IEEE Trans. Computational Intelligence and AI in Games, vol. 4, no. 1, pp. 1–43, Mar. 2012. [11] X. Liang, X. Du, G. Wang, and Z. Han, “A deep reinforcement learning network for traffic light cycle control,” IEEE Trans. Vehicular Technology, vol. 68, no. 2, pp. 1243–1253, Feb. 2019. [12] A. Cou¨etoux, J.-B. Hoock, N. Sokolovska, O. Teytaud, and N. Bonnard, “Continuous upper confidence trees,” in Proc. Int. Conf. Learning and Intelligent Optimization (LION), Rome, Italy, 2011, pp. 433–445. [13] Z. Fu, L. Wen, P. Cai, D. Fu, S. Mao, and B. Shi, “TrafficMCTS: A closed-loop traffic flow generation framework with group-based Monte Carlo tree search,” IEEE Trans. Intelligent Transportation Systems, vol. 26, no. 10, pp. 15453–15470, 2025. [14] A. Ahmad, A. S. Al-Sumaiti, Y.-J. Byon, and K. Alhosani, “Multiple intelligent control strategies for travel-time reduction of connected emergency vehicles,” IEEE Trans. Intelligent Transportation Systems, vol. 26, no. 1, pp. 337–353, Jan. 2025. [15] H. Gu et al., “Large-scale traffic signal control using constrained network partition and adaptive deep reinforcement learning,” IEEE Trans. Intelligent Transportation Systems, vol. 25, no. 7, pp. 7619–7632, Jul. 2024. [16] Highway Capacity Manual 2010, Transportation Research Board, National Research Council, Washington, D.C., 2010.

Copyright

Copyright © 2026 Tanisha Jain, Pranav Sonawane, Titeer Deshpande, Soham Lagad, Prof. B. A. Sonkamble, Prof. S. A. Joshi. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.