Dynamic Traffic Routing Problem Research Articles

Ant colony optimization-based traffic routing with intersection negotiation for connected vehicles

This study presents a decentralized traffic management system that integrates dynamic traffic routing and signal-free intersection traffic control for connected vehicles (CV). In particular, we apply the ant colony optimization algorithm and the proposed concept of colored CV to solve the dynamic traffic routing problem with multi-source multi-destination traffic flows. Each CV, modeled as an ant, deposits pheromones on its path, while road sensors aggregate the pheromone information representing traffic conditions and transfer the information to the incoming CV to perform routing. Using the colored CV concept, the CV with the same destination deposit the corresponding pheromone called colored pheromone, and only these CV can interact with that pheromone. Thus, we can distinguish multi-source multi-destination traffic flows in the routing process. For implementing signal-free intersection traffic control, we propose a decentralized intersection negotiation protocol that allows the CV to pass the intersections by exchanging their trajectory information, thereby significantly reducing the time required by CV to pass the intersections. We performed extensive simulations to evaluate our system using Simulation of Urban Mobility, a microscopic traffic simulator. The results show improvement of our proposed system over the conventional traffic management systems in terms of the mean trip time and the number of running vehicles.

Flatness Based Control for Real-Time Dynamic Traffic Routing Problem

This paper highlights the interest of differential flatness for real-time Dynamic Traffic Routing ( DTR ) problem. Such problem defined as real-time point diversion of traffic is well adapted to deal with non-recurrent congestion and in the case of major incidents. The suggested control algorithm is based on the system inversion using the concept of flatness, witch do not requires integration of any differential equations. This setting is used then for trajectories generation and tracking of the travel time function. The main objective is to find the time-dependent split variables (considered as a control variable) at the diversion point for routing the incoming traffic flow onto the alternate routes in order to reach a user equilibrium traffic pattern. A simple problem of two alternate routes demonstrates the relevance of the proposed algorithm and provides a way for generalization to more complex network with several alternate routes and multiple sections.

Time-dependent pheromones and electric-field model: a new ACO algorithm for dynamic traffic routing

In this paper, we present a dynamic ant colony optimisation (ACO) algorithm to solve dynamic traffic routing problem. The main objective of this work is to search out the least-time-cost route in a variable-edge-weight graph. We introduce time-dependent pheromones and electric-field model as two heuristic factors to improve the basic ACO. The simulation results show that the proposed dynamic ACO algorithm can effectively reduce time cost by avoiding the dynamic congestion areas. Finally, this proposed heuristic algorithm is verified to be steady-going by repeated testing.

Solution to the user equilibrium dynamic traffic routing problem using feedback linearization

In this paper, the Dynamic Traffic Routing problem is defined as the real-time point diversion of traffic during non-recurrent congestion. This dynamic traffic routing problem is then formulated as a feedback control problem that determines the time-dependent split parameters at the diversion point for routing the incoming traffic flow onto the alternate routes in order to achieve a user-equilibrium traffic pattern. Feedback linearization technique is used to solve this specific user-equilibrium formulation of the Dynamic Traffic Routing problem. The control input is the traffic split factor at the diversion point. By transforming the dynamics of the system into canonical form, a control law is obtained which cancels the nonlinearities of the system. Simulation results show that the performance of this controller on a test network is quite promising.

Fuzzy Feedback Control for Real-Time Dynamic Traffic Routing: User Equilibrium Model Formulations and Controller Design

The formulation of a system dynamics model for the dynamic traffic routing (DTR) problem is addressed, specifically for the application of real-time feedback control. Also addressed is the design of fuzzy feedback control laws for this problem. Fuzzy feedback control is suitable for solving the DTR problem, which is nonlinear and time varying and contains uncertainties. To illustrate the applicability of fuzzy logic in the design of feedback control for DTR, a simple software simulation was conducted that provided encouraging results.