Viability approach to Hamilton-Jacobi-Moskowitz problem involving variable regulation parameters

Anya Desilles
École Nationale Supérieure de Techniques Avancées

We present a few applications of the viability theory to the solution to the Hamilton-Jacobi-Moskowitz problems when the Hamiltonian (fundamental diagram) depends on time,
position and/or some regulation parameters. We study such a problem in its equivalent variational formulation. In this case, the corresponding lagrangian depends on the state of the characteristic dynamical system. As the Lax-Hopf formulae that give the solution in a semi-explicit form for an homogeneous lagrangian do not hold, we use a capture basin algorithm to compute the Moskowitz function as a viability solution of the Hamilton-Jacobi-Moskowitz
problem with general conditions (including initial, boundary and internal conditions). We present two examples of applications. In the rst one we introduce the variable speed limit as a regulation parameter. Our approach allows to compute the Moscowitz function for all values of the variable speed limit in a selected range and then to analyze its in uence on the tra
cow. In particular, we study the case when the variation of the speed limit is applied locally, in space and time.
Our second example deals with the local load capacity variations on the road. Such a variation can be a permanent property of the road (road narrowing) or it can be due to a
temporary change of number of lanes (an accident or roadworks, for example). One can also
use it as a regulation parameter by variable assignment of a supplementary lane.
1


Back to Mathematics of Traffic Flow Modeling, Estimation and Control