A stochastic adaptive control model for isolated intersections

On account of the random fluctuation of traffic demands or some special events, the signalized intersection system often exhibits severe nonlinear and time-varying behavior and therefore cannot be adequately controlled with some conventional means. A stochastic traffic signal control scheme, based on reinforcement learning, is introduced in the traffic signal control systems due to its powerful adaptability. The RL- based adaptive controller (RAC) can produced appropriate control policy to prevent the traffic network from becoming over- congested. The traditional intersection traffic model is extended to a new mode which taking some real aspects of traffic conditions into account, such as the turning fraction and the lanes scheme. The model is tested on a typical four-legged signalized intersection, and compared to both pre-timed control and full-actuated controller (FAC). Analyses of simulation results using this approach show significant improvement over traditional control, especially for the case of over-saturated traffic demand and special events such as incidents and blockages. Using the RAC model, the total mean delay of each vehicle has been reduced by 22.7% under the heavy demands compared to the FAC control algorithm.