How to compute the value of a positional differential game

From MaRDI portal





The paper deals with the computation of the value of differential games for systems described by ordinary differntial equations linear in the phase variable with a performance functional explicitly depending on the entire history of the motion. In addition to standard conditions providing the existence of the game value is assumed that the performance functional is positional and has appropriate smoothness properties. Under these assymptions a universal procedure for computing the value function of a name is presented. This procedure is based on the backward construction of upper convex hulls of auxiliary functions. The construction involves convexification only with respect to vectors dual to the phase vector of the original system for given values of an additional scalar parameter. The study model is formulated as follows. Let \[ dx/dt= A(t)x+ f(t,u,v), \quad 0\leq t_0\leq\nu,\;x\in \mathbb{R}^n,\;P\in P\subset \mathbb{R}^r,\;v\in Q\subset \mathbb{R}^s, \] be a differential system where \(x\) is the phase vector, \(u\) is the control vector (the first player's action), \(v\) is the disturbance vector (the second player's action), \(t_0\) and \(\nu\) are given time instants \((t_0<\nu)\), \(P\) and \(Q\) are given compact sets, and \(A(t)\) and \(f(t,u,v)\) are jointly continuous functions. For a differential ``small game is assumed the saddle point condition, namely \[ \min\max\langle m,f(t,u,v)\rangle= \max\min\langle m,f(t,u,v)\rangle, \quad m\in \mathbb{R}^n,\;t\in [t_0,\nu], \] where \(\langle .,.\rangle\) stands for the inner product of vectors.











This page was built for publication: How to compute the value of a positional differential game

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5951192)