1: \begin{abstract}

2: We study an optimal control problem in which both the objective

3: function and the dynamic constraint contain an uncertain

4: parameter. Since the distribution of this uncertain parameter  is

5: not exactly known, the objective function is taken as the

6: worst-case expectation over a set of possible distributions of the

7: uncertain parameter. This  ambiguity set of distributions is, in

8: turn, defined by the first two moments of the random variables

9: involved. The optimal control is found by minimizing the

10: worst-case expectation over all possible distributions in this

11: set. If the distributions are discrete, the stochastic min-max

12: optimal control problem can be  converted into a convensional

13: optimal control problem via duality, which is then approximated as

14: a finite-dimensional optimization problem via the control

15: parametrization. We derive necessary conditions of optimality and

16: propose an algorithm to solve the approximation optimization

17: problem. The results of discrete probability distribution are then

18: extended to the case with one dimensional continuous stochastic

19: variable by applying the control parametrization methodology on

20: the continuous stochastic variable, and the convergence results

21: are derived.

22:  A numerical example

23: is present to illustrate the potential application of the proposed

24: model and the effectiveness  of the algorithm.

25:  \keywords{  Distributionally robust optimal control

26:  \and Optimality necessary conditions \and Uncertainty parameter\and Duality\and Control parametrization}

27: % \PACS{PACS code1 \and PACS code2 \and more}

28: % \subclass{MSC code1 \and MSC code2 \and more}

29: 

30: 

31: \noindent {\bf AMS Subject Classification} {~~~34H05 $\cdot$ 49M25 $\cdot$ 49M37 $\cdot$ 93C41 }

32: \end{abstract}