Comparing Quadrotor Control Policies for Zero-Shot Reinforcement Learning under Uncertainty and Partial Observability | IEEE Conference Publication | IEEE Xplore