Intelligent Transportation System (ITS) has been emerged an important component and widely adopted for the smart city as it overcomes the limitations of the traditional transportation system. Existing fixed traffic light control systems split the traffic light signal into fixed duration and run in an inefficient way, therefore, it suffers from many weaknesses such as long waiting time, waste of fuel and increase in carbon emission. To tackle these issues and increase efficiency of the traffic light control system, in this work, a Dynamic and Intelligent Traffic Light Control System (DITLCS) is proposed which takes real-time traffic information as the input and dynamically adjusts the traffic light duration. Further, the proposed DITLCS runs in three modes namely Fair Mode (FM), Priority Mode (PM) and Emergency Mode (EM) where all the vehicles are considered with equal priority, vehicles of different categories are given different level of priority and emergency vehicles are given at most priority respectively. Furthermore, a deep reinforcement learning model is also proposed to switch the traffic lights in different phases (Red, Green and Yellow), and fuzzy inference system selects one mode among three modes i.e., FM, PM and EM according to the traffic information. We have evaluated DITLCS via realistic simulation on Gwalior city map of India using an open-source simulator i.e., Simulation of Urban MObility (SUMO). The simulation results prove the efficiency of DITLCS in comparison to other state of the art algorithms on various performance parameters.

Rapid growth in the domain of Internet of Things (IoT) leads to massive deployment of sensors and therefore the need to develop automatically reconfigurable complex wireless sensor network. Software defined networking (SDN) is a promising and widely adopted technique to automatic reconfigure the wireless sensor network. In SDN, control nodes are dynamically selected (to activate the functioning of the sensor network) in order to assign the tasks to other nodes and for routing data packets to control server. For residual energy of sensor nodes, and transmission distance among sensor nodes, the problem of control node selection has been formulated as an NP-hard problem. Usually, smart sensing devices of IoT suffers from low battery which are not frequently rechargeable. Therefore, an energy efficient routing mechanism is required to operate software defined wireless sensor network (SDWSN). In this paper, a green routing algorithm using fork and join adaptive particle swarm optimization (FJAPSO) is proposed to maximize the lifetime of sensor network. FJAPSO acts at two levels for auto optimization: optimal number of control nodes and optimal clustering of control nodes. Experimental results evidenced that FJAPSO outperforms other state of the art and significantly maximizes the lifetime of the sensor network.

This work proposes a traffic-light scheduling framework using the deep reinforcement learning technique to balance the traffic flow and to prevent congestion in the dense regions of the city via a software-defined control interface. A software-defined control enabled architecture is proposed to monitor the traffic conditions and it generates the traffic light control signal (Red/Yellow/Green) accordingly. For an intelligent traffic light control signal, a Deep Reinforcement Learning (DRL) model is proposed which takes vehicular dynamics as inputs from the real-time traffic environment such as heterogeneous vehicles count, speed, traffic density etc. To determine the congestion, a threshold policy is proposed and deployed on control server which generates the congestion prevention signal. A DRL agent operates in the coordination of congestion prevention signal and generates an effective traffic light control signal. The proposed model is evaluated through a realistic simulation on Indian city OpenStreetMap by using a well-known open-source simulator (SUMO). The comparative results show that the proposed solution improves several performance metrics such as average waiting time, throughput, average queue length, and average speed in the interval of 28.34% - 66.62%, 24.76% - 66.60%, 30.89% - 69.80%, and 16.62% - 43.67% respectively over other states of the art approaches.

Traffic congestion is a pressing issue worldwide, and machine learning (ML) methods are increasingly being used in Intelligent Transportation Systems (ITS) to address this problem. Deep hybrid models, in particular, have emerged as an efficient solution for traffic flow prediction. Among these models, Long Short-Term Memory (LSTM) and Bidirectional LSTM (Bi-LSTM) have been widely used to capture the temporal and periodic features of traffic data. The advancements in RNNs provide an opportunity to enhance the performance of existing models. Therefore, this work proposes a BiGRU-BiGRU model with two modules to extract temporal and periodic features from traffic data. Recurrent Neural Networks (RNNs) have proven to perform well using attention mechanism. However, there is a need for attention mechanism that strictly focuses on traffic dynamics and nearby data from the prediction points. Thus, a novel confined attention mechanism is proposed and incorporated into the first module to improve the model’s performance by focusing only on the recent relevant information in the traffic flow sequence. Furthermore, the external features are integrated to improve the model’s prediction performance. The proposed model is evaluated on the publicly available real-world dataset and compared with several baseline state-of-the-art methods. As an outcome, the model offers a reduction in the average value of RMSE, MAE, and MAPE for all the prediction horizons, that is ranged from 5.1% – 20.4%, 7.3% – 27.3%, and 6.1% – 56.6%, respectively.