Keywords

1 Introduction

Nowadays, traffic congestion becomes a big problem. Therefore wastage of time and fuel are increasing for the travelling vehicles due to the increase of traffic intensity. Traffic efficiency is one of the main applications that have been developed over road networks using VANETs. Not only is this but a huge amount of pollution is also occurred. To preserve the time and fuel and control the pollution, the congestion control of the traffic in roadways can take a major role. There are several ways to reduce the congestion. One of the schemes is traffic light scheduling. Especially in the city, the junction of the road networks is crowded with vehicles. Because, the vehicles need to wait to cross the roads in the junction points. But in the peak hours the number of vehicles becomes very high. They need to wait for long time to cross the road. Traffic lights control the traffic flows at each road intersection. They provide safe scheduling that allows all conflicting traffic flows to share the road intersection. Several research works is going on in this area. Optimization of dynamic traffic light scheduling scheme is made in such a way so that the traffics, waiting in the junction should not wait for unnecessarily long time. The scheme increases the traffic flow in the roadways and minimizes the waiting time for the traffic. The queuing delay at each signalized road intersection decreases traffic fluency, which decreases traffic efficiency throughout the road network. To enhance the performance of traffic efficiency, several researchers have developed intelligent and efficient algorithms to schedule the increase the flow of traffic at each signalized road intersection. The efficient schedule for each traffic light should reduce the waiting delay time of traveling vehicles at each road intersection and increase the throughput of road intersections.

In this paper, an intelligent traffic light controlling algorithm is introduced considering a real-time traffic characteristics of all traffic flows that will cross the road intersection. The introduced algorithm is intended to schedule the traffic flows at each road intersection, while reducing the expected queuing length and waiting time of vehicle and increasing the throughput of the road intersection. A virtual ready area has been considered around each road intersection. Vehicles inside the boundaries of this area are ready to cross the intersection. However, the vehicles receive a red signal from the traffic light begin to decrease their speed and prepare to stop at the closest empty space to the road intersection. Similarly, the vehicles receive green signal from traffic light, start the engine, accelerate the speed and exit the road area. The flows of traffic are density is scheduled to pass first. Moreover, the assigned time of each phase of the timing cycle is set based on the real-time traffic distribution inside the ready area.

2 Related Work

In this section we briefly discuss about the related research work in this field. Intelligent Traffic Light Controlling (ITLC) [1] algorithm is proposed considering the real-time traffic characteristics of each traffic flow are going to cross the road junction. The methodology proposed here for scheduling the time phases of each traffic light. A new approach [2] for smart traffic light control at intersection is proposed. A connected intersection system where every object like vehicles, sensors, and traffic lights all are connected and sharing information to one another. The controller is able to collect effectively and mobility traffic flow at intersection in real-time. In the work [2] authors also propose the optimization algorithms for traffic lights by applying algorithmic game theory. Two game models (Cournot Model and Stackelberg Model) are used to deal with difference scenarios of traffic flow. In this regard, based on the density of vehicles, controller will make real-time decisions for the time durations of traffic lights to optimize traffic flow. In [3], an algorithm is proposed for system to control the traffic by measuring the realtime vehicle density using canny edge detection with digital image processing. The procedure offers traffic control system. Besides that, the complete technique from image acquisition to edge detection and green signal allotment using sample images of different traffic conditions is depicted. A study, inspired by recent advanced vehicle technologies [4], considering for improvement of traffic flow in real-time problem. The algorithm depicts a new approach to manage traffic flow at the intersection by scheduling of traffic light signal. The method is based on process synchronization and connected vehicle technology. The traffic deadlock is also considered for huge number of traffic. The simulation shows the potential results comparing with the existing traffic management system. A new approach [5] is proposed for traffic flow management at intersection. By IoT, based on connected object, A model is designed which communicating among objects to improve traffic flow at intersection with real time problem. In this scheme, traffic congestion is also considered in case of high traffic volume. Traffic Congestion Investigating System by Image Processing [6] from CCTV Camera is proposed to check a traffic condition from a traffic image on road. The system brings a traffic image from a CCTV camera to process in the system as an input. Then, the system finds for traffic congestion and gets the results in three traffic conditions as Flow, Heavy, and Jammed. Finally, a user can use the system for a transportation planning or an intersection traffic control. In this paper [7], a new traffic system recommendation based on support real-time flows in highly unpredictable sensor network environments is proposed. The proposed algorithm includes two phases. First phase is proposed to deal with the real-time problem and second phase, the algorithm is based on Depth First Search (DFS) algorithm to recommend the paths which meet demands of drivers based their context. A unique Intelligent Road Traffic Monitoring And Management System [8] is proposed to improve traffic flow and safety of road users. An Intelligent Road Traffic Monitoring and Management System (IRTMMS) [8] based on the VANET is proposed in this paper. A new scheme [9] proposes that reduces the traffic congestion problems and improves the performance decreasing the vehicle waiting time and their pollutant emissions at intersections. Combination of vehicle-to-vehicle (V2V) and vehicle-to-infrastructure (V2I) communications is used to fulfill the goal. The main traffic input for applying traffic assessment in this approach is the queue length of vehicle clusters at the intersections. A new approach is proposed [10] where road-side facilities communicate the traffic light cycle information to the approaching vehicles. Based on this information, the vehicles determine their optimal speeds and other appropriate actions to take to cross road intersections with minimum delays. The survey [11] of adaptive signal control strategies to optimize traffic signal is done where authors divided the procedure in three categories depending on level of traffic involvement. Two secure intelligent traffic light control schemes [14] are done using fog computing whose security are based on the hardness of the computational Diffie–Hellman puzzle and the hash collision puzzlerespectively. Designing of a dynamic and efficient traffic light scheduling algorithm [15] that adjusts the best green phase time of each traffic flow, based on the real-time traffic distribution around the signalized road intersection. This proposed algorithm has also considered the presence of emergency vehicles, allowing them to pass through the signalized intersection as soon as possible. The phases of each traffic light are set to allow any emergency vehicle approaching the signalized intersection to pass smoothly. In case of multiple emergency vehicles approach the signalized intersection have been investigated to select the most efficient and suitable schedule. A survey [16] of different traffic density estimation methodologies is described, where one can get the vivid idea about the trends of research work in this aspect. A multi-agent traffic signal control system based on vehicular ad hoc network (VANET) is proposed [17] where real-time and accurate vehicle information obtained by vehicular ad hoc network is used. By constructing a distributed architecture using multi-agent technology, and realizing vehicle road communication using communication module of intersection agent, a more intelligent signal timing strategy is constructed by discrete fuzzy controller. A novel preemptive algorithm for optimization of traffic signals in VANETs [18] proposes to reduce large queuing at crossroads by allowing smooth movement of traffic on the roads without much waiting. The proposed algorithm selects green light timings according to real time vehicular density and can select any phase out of predefined order depending on the traffic density on that phase to reduce the congestion. A new scheme for dynamic traffic regulation method is proposed based on virtual traffic light (VTL) [19] for Vehicle Ad Hoc Network (VANET). In our framework, each vehicle can express its “will”—the desire of moving forward—and share among one another its “will”-value and related traffic information at a traffic light controlled intersection. A decentralized model based on multi-agent systems (MAS) and complex event processing (CEP) [20] is proposed. The new control scheme improves green light time to reduce the average waiting time of vehicles. This improvement is provided by the observation of the intersection through Cyber Physical Systems (CPS). The paper [21] proposes an auto-adaptive model for smart regulation traffic lights. The research work provides an intelligent traffic light control system to avoid the traffic congestion and to give a free way to emergency vehicles to reach their respective places without any delay. This system uses IR sensors to detect the vehicles density before signal. This will help the Arduino to change the signal timing based on the number of vehicles. For traffic clearance, the radio frequency transmitter and receiver are used. To detect the emergency vehicles and will change the signal color from red to green. Thus, this intelligent traffic light control system will help us to avoid traffic congestion and provide traffic clearance for emergency vehicles to reach their destination.

3 Proposed Methodology

Phase: A phase is a set of routes of vehicles. The vehicles of all the routes in the same phase can cross the junction simultaneously without colliding each other (Figs. 1, 2, 3 and 4).

Fig. 1.
figure 1

Phase 1

Full size image
Fig. 2.
figure 2

Phase 2

Full size image
Fig. 3.
figure 3

Phase 3

Full size image
Fig. 4.
figure 4

Phase 4

Full size image

Row of Vehicles:

A row of vehicles is defined as the number of vehicles standing beside each other (perpendicularly to the road) during the red signal.

Queue length of a route = number of vehicles in a route.

Queue length of a phase = summation of number of vehicles in all the routes of a phase.

Queue length of a lane = number of vehicles in all the routes of a lane (Fig. 5).

Fig. 5.
figure 5

Row of vehicles for route and lane

Full size image

Assumption: A lane has three routes to follow. The routes are right turn route, straight route and left turn route. The right turn route is always open. So, in present work, only straight route and left turn route are considered for traffic light scheduling.

3.1 Tasks of RSUs

  • Receives message from each vehicle of all the routes.

  • Calculates number of vehicles for each route of all phases from the received message of vehicles.

  • Calculates queue length of a route as the number of vehicles of the route.

  • Calculates summation of queue length of two routes of each phase.

  • Calculates sub set of queue length as the difference between the queue length of the selected phase and next highest queue length phase.

  • Calculates sub set of queue length of highest queue length phase as the difference between the queue length of the highest queue length phase and next highest queue length phase.

  • Calculates row of vehicles for highest queue length phase as the number of vehicles standing beside each other (perpendicularly of a road).

  • Counts number of rows of vehicles for selected subset of highest queue length phase as the ratio of number of vehicles of the subset of queue length and number of vehicles of each row.

  • If total number of vehicles of the routes in a phase is pj_tot_sub_vehi, (where j denotes the phase number and i denotes the route number) and the number of vehicle of a row is row_vehi,j where i and j denotes the route and phase respectively. For a particular phase pj

  • Number of rows = \( (\sum\nolimits_{{{\text{i}} = 1}}^{\text{n}} {{\text{p}}_{\text{j}} \_{\text{tot}}\_{\text{sub}}\_{\text{veh}}_{\text{i}} )/{\text{row}}\_{\text{veh}}_{\text{i,j}} } \)

  • Calculates duration of green signal for highest queue length phase as the time required for a sub set of queue length of highest phase to cross the junction.

  • Calculates the duration of red signal for other phases as equal like the duration of green signal.

  • Sends duration of green signal for highest queue length phase and red signal for other phases to Traffic Management Controller (TMC).

  • Calculates congestion of each incoming lane having red signal as the distance of the last row of vehicle from the junction. The distance is calculated as the summation of all the inter row distance of the lane and distance of first row from the junction.

  • Compare congestion of incoming lane with a predefined threshold of that lane and sends a message to TMC in case the congestion of lane crosses the threshold.

  • The threshold value is calculated as-

  • Congestion threshold = length of the lane/2.

  • Repeats the same steps of operations.

3.2 Tasks of TMC or VANET Authority

  • Receives green signal duration of one phase and red signal duration of other phases from RSUs.

  • Assigns green signal for one phase and red signal to other phases as per the information of RSUs.

  • Receives congestion message of congested lane from RSUs (if congestion occurred).

  • Increases the waiting time of the routes of connecting junction coming towards the congested lane (if congestion occurred).

  • Calculates the increment of waiting time as (duration of green signal of the congested lane)/2 (Table 1).

    Table 1. Different notations used in the algorithm 1
    Full size table

3.3 Proposed Algorithm

(See Algoritm 1)

figure a

3.4 Flow Chart

(See Fig. 6)

Fig. 6.
figure 6

Flow chart for proposed methodology

Full size image

3.5 Performance Analysis

To evaluate the algorithm, a network of six intersections (4 point intersection) is taken for consideration. Different number of vehicles is taken as input of the network for various times. For simulation, SUMO 0.25.0 is used for creating the network and the vehicle movement. The communication with RSU is done by omnetpp 4.9. The parameters taken for comparison are average queuing length of vehicles in a particular intersection and the average waiting time of the vehicles at the end of the simulation. The comparison graphs are given below. It is assumed that average speed of the vehicles is 60 km/hr, acceleration speed is 6 km/hr2 and start time is 20 s, inter vehicular distance is 2 m. Figs. 7 and 8 depicts the comparison graph with the proposed method and other two methods (one is Xiao et al. (2017) and Punam et al. (2015)). The proposed method gives better result than that of other two methodologies Xiao et al. (2017) and Punam et al. (2015).

Fig. 7.
figure 7

No. of vehicles vs. average queuing length

Full size image
Fig. 8.
figure 8

No. of vehicles vs. average waiting time

Full size image

4 Conclusion

The research work done in this paper is concentrating on traffic light scheduling in urban area road network. A methodology in order to optimize the traffic light scheduling is proposed to control the road traffic congestion in urban road network. The proposed methodology is established by the implementation of a new scheduling algorithm. The proposed method targets to reduce the queuing length of road traffic and their average waiting time. The algorithm is simulated on SUMO 0.25.0 and OMNETPP 4.9. The simulated result is compared with other methodologies. The simulation result shows better performance depending upon the aforementioned parameter.