Adaptive Deep Q-Network Algorithm with Exponential Reward Mechanism for Traffic Control in Urban Intersection Networks

Abstract

The demand for transportation has increased significantly in recent decades in line with the increasing demand for passenger and freight mobility, especially in urban areas. One of the most negative impacts is the increasing level of traffic congestion. A possible short-term solution to solve this problem is to utilize a traffic control system. However, most traffic control systems still use classical control algorithms with the green phase sequence determined, based on a specific strategy. Studies have proven that this approach does not provide the expected congestion solution. In this paper, an adaptive traffic controller was developed that uses a reinforcement learning algorithm called deep Q-network (DQN). Since the DQN performance is determined by reward selection, an exponential reward function, based on the macroscopic fundamental diagram (MFD) of the distribution of vehicle density at intersections was considered. The action taken by the DQN is determining traffic phases, based on various rewards, ranging from pressure to adaptive loading of pressure and queue length. The reinforcement learning algorithm was then applied to the SUMO traffic simulation software to assess the effectiveness of the proposed strategy. The DQN-based control algorithm with the adaptive reward mechanism achieved the best performance with a vehicle throughput of 56,384 vehicles, followed by the classical and conventional control methods, such as Webster (50,366 vehicles), max-pressure (50,541 vehicles) and uniform (46,241 vehicles) traffic control. The significant increase in vehicle throughput achieved by the adaptive DQN-based control algorithm with an exponential reward mechanism means that the proposed traffic control could increase the area productivity, implying that the intersections could accommodate more vehicles so that the possibility of congestion was reduced. The algorithm performed remarkably in preventing congestion in a traffic network model of Central Jakarta as one of the world’s most congested cities. This result indicates that traffic control design using MFD as a performance measure can be a successful future direction in the development of reinforcement learning for traffic control systems.

A Resilient Intelligent Traffic Signal Control Scheme for Accident Scenario at Intersections via Deep Reinforcement Learning

“…This state definition has resulted in the desired performance in low traffic volume scenarios. A reinforcement learning method was proposed in [37] for achieving maximum intersection throughput. Based on the total pressure and the total queue length at an intersection, [37] defines an adaptive reward function that uses an exponential approach.…”

“…A reinforcement learning method was proposed in [37] for achieving maximum intersection throughput. Based on the total pressure and the total queue length at an intersection, [37] defines an adaptive reward function that uses an exponential approach. Recently, safety and accident analysis have become increasingly popular in traffic signal control [38][39][40][41].…”

An efficient algorithm for optimal route node sensing in smart tourism Urban traffic based on priority constraints

“…Set the scale of road network negotiation from small to large and choose four (1,2,5,6), nine (1,2,3,5,6,7,9,10,11) and sixteen (16) junctions to observe the performance of the proposed model in various sized road networks. A, B, C, and D are utilized as source addresses and E, F, and G are used as destinations in the simulation experiment.…”

“…• Control the dynamic optimization of traffic signals using technologies like SCOOT (split cycle offset optimization method) [2] and SCATS (sydney coordinated adaptive traffic system) [3]. This kind of approach primarily relieves traffic congestion from the standpoint of macro-traffic ŕow and increases the rate at which road resources are utilized by accelerating the circulation speed of driving vehicles, but it ignores the various traffic requirements of each microscopic driving vehicle; • Deterministic algorithms, such as the A* algorithm [4], the Dijkstra shortest path algorithm [5], the dynamic programming method [6], etc., make up the majority of the path navigation algorithms for a single moving vehicle. PSO algorithm [7], genetic algorithm [8], colony optimization algorithm [9], neural network algorithm [10], and other intelligent algorithms.…”