Advanced Control Strategies For Unban Freeway Traffic Flow

Urban freeways had been originally designed to provide virtually unlimited mobility to road users. The on-going dramatic expansion of car-ownership, however, has led to the daily appearance of freeway congestions. It has been recently realized that the mere infrastructure expansion cannot provide a complete solution to these problems due to economic and environmental reasons or, in metropolitan areas, simply due to lack of space. So various control strategies combined powerful control technique have been proposed and employed to ameliorate traffic conditions on freeway. Ramp metering that regulates the vehicles to the freeway from the entry ramp, route guidance that provide the optimal route to the drivers and the speed limitation that improves the stability of the traffic system have been widely investigated. However due to the complex and stochastic characteristics of traffic dynamic system, so far no a policy has been regarded perfect and adequate to the freeway control. Ramp metering being an essential method to the freeway control becomes the basis of our work.In this dissertation to begin with a comprehensive overview of existing ramp metering algorithms is provided. These algorithms were classified into the different categories depend on the various criteria such as the respondence to the real-time, the controlled system and information manage. From these algorithms we know which category is effective and what is the attracting ideas of this algorithm to be borrowed. On the basis of these algorithms, three advanced strategies for different traffic system are proposedFirst the robust control approach for the local ramp metering is proposed based on the famous macroscopic traffic LRW theory. Compared with the optimal LQR control, the robust control considers the influence of the unpredicted and uncontrolled disturbance to traffic system. For this reason, the simulation results show the superiority of the robust control to the optimal control for the changeful traffic conditions.Next a new hierarchy control strategy implied for coordinated ramp metering is presented. Depending on the system wide traffic information this control policy tries to perform adequate in the various traffic situations. The three control layers are the direct control layer for the local ramp metering to benefit the controlled freeway section, the coordinated control layer via adaptive fuzzy control considering themulti-ramp traffic and the adaptive layer including predicting possible congestion, updating the parameters of the coordinated algorithm and the adaptive decisions. To evaluate the efficiency and feasibility of the proposed control strategy, the detail design for the controller involving the predictive algorithms to foresee the traffic future condition, migration genetic algorithm to tune parameters of fuzzy controller, and the optimal control to local ramp metering is implemented to the loop freeway system.Finally the integrated control strategy framework is discussed to manage the freeway corridor system. The five control measures: variable speed limitation, ramp metering, route guidance exit freeway advice and coordinated intersection signal control are integrated to the different degree responding to the different traffic situations.