A novel approach for dynamic traffic lights management based on Wireless Sensor Networks and multiple fuzzy logic controllers

• Flexible, scalable traffic lights dynamic control system for isolated intersections.
• Multiple parallel fuzzy controllers dynamically manage both the phase and green time.
• Outperforms other works in terms of vehicle waiting time and balancing between phases.
• Very effective under heavy traffic and when phases have unbalanced arrival rates.
• Lightweight effective implementable on COTS devices: broad practical impact expected.

This paper proposes a novel approach to dynamically manage the traffic lights cycles and phases in an isolated intersection. The target of the work is a system that, comparing with previous solutions, offers improved performance, is flexible and can be implemented on off-the-shelf components. The challenge here is to find an effective design that achieves the target while avoiding complex and computationally expensive solutions, which would not be appropriate for the problem at hand and would impair the practical applicability of the approach in real scenarios. The proposed solution is a traffic lights dynamic control system that combines an IEEE 802.15.4 Wireless Sensor Network (WSN) for real-time traffic monitoring with multiple fuzzy logic controllers, one for each phase, that work in parallel. Each fuzzy controller addresses vehicles turning movements and dynamically manages both the phase and the green time of traffic lights. The proposed system combines the advantages of the WSN, such as easy deployment and maintenance, flexibility, low cost, noninvasiveness, and scalability, with the benefits of using four parallel fuzzy controllers, i.e., better performance, fault-tolerance, and support for phase-specific management. Simulation results show that the proposed system outperforms other solutions in the literature, significantly reducing the vehicles waiting times. A proof-of-concept implementation on an off-the-shelf device proves that the proposed controller does not require powerful hardware and can be easily implemented on a low-cost device, thus paving the way for extensive usage in practice.