Smart traffic management system through optimized Network Architecture for the smart city paradigm shift

A core premise of smart city projects is the use of technology to better manage assets and resources. City transportation infrastructure is typically a prominent concern. The future generation of traffic systems in a connected world will offer a thorough traffic information framework to support all modes of transportation, including cars, public transportation, ambulances, containers, trucks, bicycles, and pedestrians. Existing detection systems provide fundamental information such as vehicle count, occupancy, and transit data. Contemporary traffic control management systems allow one to keep an eye on signal operations, update signal controls and plans based on the time of day or traffic volume, and even provide adaptive signal timing, which modifies signal timing parameters based on traditional car detector data. This project provides comprehensive system optimization to enable transit and signal priority, emergency vehicle preemption, and pedestrian movements while improving overall network performance. There is an increasing demand for maximum network bandwidth and high-availability communications as smart transportation systems, connected vehicle technologies, and autonomous vehicles approach. To correctly estimate the phase of a particular signal, information on vehicle presence, vehicle count, and delay time is gathered. The amount of time a vehicle must wait before moving on following an assault can be used to gauge its importance. The model will utilise the SUMO data on vehicle presence to decide which signal phase to send as an output in order to reduce the cumulative time delay for automobiles at any given instant. The reinforcement algorithm will decide if the output has produced a desirable situation and will deliver arbitrary rewards as output. The learning algorithm will anticipate which phase to transition to. The cumulative vehicle delay time, which is the time required to arrive at a given destination and vehicle queue length, has decreased once the DQN algorithm has been used in place of the round-robin approach.