No label defined (Q3972977)

[For the entire collection see Zbl 0719.00020.] This paper provides a theory of necessary conditions for a class of dynamic optimization problems: \[ \text{Minimize } \sum^ N_{i=1} \int^{t^ +_ i}_{t^ -_ i} \ell_ i(t,x_ i(t), u_ i(t))dt+h(\{t^ -_ i,t^ +_ i, x_ i(t^ -_ i), x_ i(t^ +_ i)\}), \] subject to \(\dot x_ i=\phi_ i(t,x_ i(t), u_ i(t))\), \(x_ i(t)\in X_ i(t)\), \(u_ i(t)\in\Omega_ i(t)\), a.e. \(t\in[t^ - _ i,t^ +_ i]\), for \(i=1, 2,\dots,N\), and \(\{t^ -_ i,t^ +_ i, x_ i(t^ -_ i), x_ i(t^ +_ i)\}^ N_{i=1}\in\Lambda\), where \(\ell_ i\), \(f_ i\) and \(h\) are functions and \(\Omega_ i(t)\) and \(\Lambda\) are given sets. This is a nonstandard optimization problem. It involves a family of differential equations with control term, coupled through the cost function and the endpoints constraints. It is called an optimal multiprocess problem. A maximum principle governing the optimal solution was derived without proof.