The motion planning of powered descent problems has often been treated in the deterministic optimal control framework, which provides efficient theoretical and numerical tools. However, future applications require robustness, usually obtained by introducing stochastic components in the dynamics to model uncertainties. After stating the robust motion planning problem, this paper proposes a deterministic approximation which avoids the computational difficulties of stochastic optimal control. The approach consists of guiding the mean while reducing the covariance, the dynamics of these two quantities being approximated thanks to statistical linearization. In addition, since feedback control is necessary to control covariance, two techniques are provided to deal with actuator limits when the control is stochastic.