A Notation

Table 1 provides an overview of the notation used in this paper. In this section we will consider first the more general case where the state is partially observable, and derive the equations for a fully-observable state as a special case afterwards. Eq. (18) gives directly the formula for the approximation of p(x If not stated otherwise, we used 100 trajectories to determine the MLEs. In this section we will provide additional results for the reaching and random problems. Figure 1 shows the RMSEs of our method in comparison to the two baselines described in Section 4.1. Evaluation of the proposed inverse optimal control method in terms of RMSE of the maximum-likelihood parameter estimates.