Flow based local ramp metering using iterative learning and PARAMICS platform

In this work, we apply iterative learning control to address ramp metering in a macroscopic level freeway environment. By formulating the original ramp metering problem as a output tracking and disturbance rejection problem, iterative learning control can be applied and improve the traffic response. The learning mechanism is based on traffic flow. To mimic a real traffic environment, a simulator is built on the PARAMICS platform. The control design and simulations are carried within the platform under on-ramp constraints. The effectiveness of the new approach is verified through PARAMICS-based simulations.