Coordinated freeway ramp control based on genetic ant colony algorithm

In this paper, we employ a multilayer decentralized control and an advanced PID control to solve the coordinated ramp control problem. Furthermore, we use a hybrid optimization approach to adjust the PID parameters. This hybrid approach in use is the combination of a genetic algorithm (GA) and an ant colony algorithm (ACA). First, a macroscopic and dynamic freeway traffic flow model called a finite-difference model is built. Second, the control objective of ramp metering is fully set forth. Third, a coordinated control system of multiple ramps based on a genetic ant colony algorithm is designed. This control system is divided into an adaptation layer, a coordination layer, and a direct control layer. In the direct control layer, we employ PID control to determine the ramp metering rates and use a hybrid algorithm of GA and ACA to optimize the PID parameters. Finally, simulation research and result comparison are carried out. Simulation results show that the method of multilayer decentralized control, as well as the hybrid optimization algorithm, has an excellent density tracking performance. This control technology, as well as the optimization process, finds a new way for freeway traffic control.