Iterative adaptive compensation of modeling uncertainties in emission control of freeway traffic

摘要

Control of the emission rate of exhaust fumes as well as vehicle density in various roadsegments of freeway traffic practically is desirable. For this purpose hydrodynamicmacroscopic models are used that approximate the traffic as the motion of some compressiblefluid. They normally have analytical ambiguities due to using various finite elementapproximations of spatial partial differential operators and applying various analyticalcompressibility models. Further uncertainties stem from the estimation of the modelparameters. The suggested approach is able to iteratively and adaptively refine the trafficcontrol based on any version of the possible analytical forms. The controller has to measurevehicle densities and traffic velocities and applies variable traffic signs for prescribingvelocities and allowed ingress rate from a ramp as control signals. Since the stationarysolutions of the control problem are stable in the region investigated, instead of fast, reallydynamic control it applies simple quasi-stationary process model and control that is acommon practice in Classical Thermodynamics. The operation of the proposed method isillustrated via simulations.